Ese

IGa

b

c

a

ARA

KITBI

1

tbtptitp(

h0

Molecular Immunology 60 (2014) 1–7

Contents lists available at ScienceDirect

Molecular Immunology

j ourna l ho me pa ge: www.elsev ier .com/ locate /mol imm

xcreted/secreted Trichuris suis products reduce barrier function anduppress inflammatory cytokine production of intestinalpithelial cells

.H. Hiemstraa,1, E.J. Klavera,1, K. Vrijlanda, H. Kringelb, A. Andreasenb,. Boumac, G. Kraala, I. van Diea,∗,1, J.M.M. den Haana,1

Department of Molecular Cell Biology and Immunology, VU University Medical Center, Amsterdam, The NetherlandsSection for Parasitology, Health and Development Department of Veterinary Disease Biology, University of Copenhagen, DenmarkDepartment of Gastroenterology, VU University Medical Center, Amsterdam, The Netherlands

r t i c l e i n f o

rticle history:eceived 26 February 2014ccepted 10 March 2014

eywords:ntestinal epithelial cellsrichuris suisarrier function

mmune therapy

a b s t r a c t

The administration of helminths is considered a promising strategy for the treatment of autoimmunediseases due to their immunomodulatory properties. Currently, the application of the helminth Trichurissuis as a treatment for Crohn’s disease is being studied in large multi-center clinical trials. The intestinalepithelium forms an efficient barrier between the intestinal lumen containing the microbial flora andhelminths, and dendritic cells (DCs) present in the lamina propria that determine the TH response. Here,we investigated how excreted/secreted (E/S) products of T. suis affect the barrier function of intestinalepithelial cells (IECs) in order to reach the DCs and modulate the immune response. We show that T.suis E/S products reduce the barrier function and the expression of the tight junction proteins EMP-1and claudin-4 in IEC CMT93/69 monolayers in a glycan-dependent manner. This resulted in an increasedpassage of soluble compounds to the basolateral side that affected DC function. In addition, T. suis E/Ssuppressed LPS-induced pro-inflammatory cytokine production by CMT93/69 cells, whereas the pro-

duction of the TH2 response-inducing cytokine thymic stromal lymphopoietin (TSLP) was induced. Ourstudies indicate that T. suis E/S glycans affect the function of the intestinal epithelium in order to mod-ulate DC function. Identification of the T. suis E/S glycans that modulate IEC and DC function may leadto a strategy to reduce symptoms of autoimmune and allergic immune diseases by orally administratedhelminth-derived factors without the need of infection with live helminths.

. Introduction

The increased incidence of autoimmune and allergic diseases inhe industrialized western world has been proposed to be causedy the decrease in infections with certain microorganisms, in par-icular helminths, due to improved hygiene. Indeed, intestinalarasitic infection is associated with reduced risk of skin sensitiza-ion to allergens (Feary et al., 2011), and anti-helminth treatmentncreases reactivity of the skin to allergens (Flohr et al., 2010). Forhis reason administration of helminths or helminth products is

roposed to prevent or treat autoimmune and allergic diseasesJouvin and Kinet, 2012).

∗ Corresponding author. Tel.: +31 20 444 8157; fax: +31 20 444 8081.E-mail address: im.vandie@vumc.nl (I. van Die).

1 These authors contributed equally to this work.

ttp://dx.doi.org/10.1016/j.molimm.2014.03.003161-5890/© 2014 Elsevier Ltd. All rights reserved.

© 2014 Elsevier Ltd. All rights reserved.

The helminth Trichuris suis belongs to the nematode family andis a natural parasite of pigs. Ingested fertilized eggs hatch in theduodenum, where they release larvae that develop into worms thatmigrate to the caecum and proximal colon. The worms secrete solu-ble products (excretory/secretory products, E/S) that can modulateand/or skew immune responses (Hewitson et al., 2009). AlthoughT. suis is a pig whipworm, it has been shown to shortly colo-nize the human intestine, without inducing any adverse symptoms(Beer, 1976). Oral administration of T. suis eggs has given promis-ing results in initial small clinical trials inducing disease remissionin patients suffering from Crohn’s disease, ulcerative colitis, andmultiple sclerosis (Summers et al., 2005a,b; Fleming et al., 2011).

The effectiveness of T. suis administration in diminishingsymptoms of such inflammatory diseases can be explained by

the type of immune response that is induced by this helminth.Whereas these diseases in general are driven by a TH1- and/orTH17 mediated immune response, helminths stimulate the devel-opment of a TH2-dominated immune response (Allen and Maizels,

2 ular Im

2it(

Tptpi2Trr2wipniue

faibatEprtstna

2

2

HbF5sIn

(T(Fo

2

osa(w

I.H. Hiemstra et al. / Molec

011). TH2-induced cytokines together with helminth-mediatednduction of regulatory T cells suppress TH1- and TH17 responseshat drive the disease and as a result disease symptoms are reducedSummers et al., 2005b; Elliott et al., 2003).

Dendritic cells (DCs) are the main cell type to activate naïve cells and determine the type of T cell response, in part viaroduction of cytokines that drive TH1 or TH2 responses. Similaro other helminths such as Schistosoma mansoni, T. suis solubleroducts were shown to suppress the production of several pro-

nflammatory cytokines and chemokines by DCs (Pearce et al.,004; Klaver et al., 2013; Kuijk et al., 2012), among which isH1-inducing cytokine interleukin-12 (IL-12) (Kuijk et al., 2012),esulting in a T. suis-dependent reduction of TH1 and TH17esponses, while stimulating TH2 cell differentiation (Kuijk et al.,012). However, it is not clear whether the presence of T. suisorms in the intestines can also affect other cell types and thereby

nfluence immune responses, and how T. suis can modulate a DCopulation that is present in the lamina propria. Since the intesti-al epithelium forms a physical barrier that prevents entry of the

ntestinal microflora, we set out to investigate whether E/S prod-cts that are naturally released by T. suis in its environment, affectpithelial barrier function.

Here we show that T. suis E/S products reduce the barrierunction of monolayers of intestinal epithelial CMT93/69 cells in

glycan-dependent manner, which coincided with a reductionn gene expression of tight junction molecules epithelial mem-rane protein 1 (EMP-1) and claudin-4. T. suis E/S products passnd/or modulate the CMT93/69 monolayer to suppress inflamma-ory cytokine production by DCs. In addition, we show that T. suis/S products suppress pro-inflammatory cytokine and chemokineroduction by CMT93/69 cells, whereas the production of TH2esponse-inducing cytokine TSLP was induced. These data showhat T. suis exerts direct effects on intestinal epithelial cells, byuppressing pro-inflammatory cytokine and chemokine produc-ion and by increasing barrier permeability. These data provideovel insights into the mechanisms by which T. suis E/S productsffect the intestinal epithelium to modulate DC functions.

. Materials and methods

.1. Epithelial cell culture

Murine colorectal epithelial CMT93/69 cells (Franks andemmings, 1978) were cultured on collagen-coated 24 wells flatottom plates in DMEM supplemented with 10% heat-inactivatedCS, 1% l-glutamine, and 1% penicillin–streptomycin at 37 ◦C and% CO2. After 3 days, confluent monolayers were stimulated with T.uis E/S and/or ultrapure LPS (Escherichia coli 0111:B4; 100 ng/ml,nvivogen). Cells were tested regularly for mycoplasma contami-ation.

Permeability of monolayers grown on transwell membranes0.4 �M pore size, Costar, Corning Life Sciences BV, Amsterdam,he Netherlands) was determined by translocation of FITC-dextran1 mg/ml) from the apical to the basolateral side. The fluorescentITC-signal in the basolateral medium was measured using a Flu-star spectrofluorimeter (BMG Labtec, Offenburg, Germany).

.2. Preparation of helminth excreted/secreted proteins (E/S)

Adult T. suis worms were collected from the caecum and colonf infected pigs and washed extensively with 0.98% NaCl. The live T.

uis worms were incubated with RPMI1640 containing 1% glucosend penicillin (500 U/ml), streptomycin (0.5 mg/ml) anf fungizone1.25 �g/ml) for two days at 37 ◦C. Incubation media were replacedith fresh RPMI media every day and the collected supernatants

munology 60 (2014) 1–7

were filtered through a 0.45 �M filter. The filtrate was concen-trated using Amicon-Ultra-15 10 kDa centrifugal filters (Millipore).The E/S fraction containing proteins larger than 10 kDa was filteredthrough a 0.2 �M filter. A BCA assay (Pierce) was performed todetermine the T. suis E/S protein concentration.

To denature protein structures, T. suis E/S was incubated for20 min at 80 ◦C, then immediately cooled on ice and stored at −20 ◦Cuntil use. Further breakdown of proteins was achieved by addi-tional overnight incubation at 37 ◦C with the proteolytic enzyme�-chymotrypsin (Sigma, USA) (1 mg/ml). The �-chymotrypsinactivity was heat inactivated for 20 min at 60 ◦C.

To oxidize glycan moieties, T. suis E/S was treated with sodiumperiodate (Sigma, USA, 10 mM) at pH 5.0, for 45 min at room tem-perature (RT) in the dark. Subsequently, reducing agent sodiumborohydride (Merck, Germany, 50 mM) was added for 25 min atRT in the dark. To remove periodate and borohydride, the samplewas dialysed against PBS (Lonza, Switzerland, 106-fold excess) for20 h, using 2 K MCWO Slide-A-Lyzer cassettes (Pierce, USA).

2.3. Monocyte isolation and generation of immature dendriticcells

Human monocytes were isolated from buffy coats of healthydonors (Sanquin, Amsterdam, The Netherlands). All blood donorsgave informed consent. Monocytes were isolated from PBMCs usingCD14 microbeads (MACS, Miltenyi Biotec, Germany), as previouslydescribed (van Stijn et al., 2009) and differentiated into immaturedendritic cells (DCs) in the presence of 250 IU/ml IL-4 and 120 IU/mlGM-CSF (Immunotools GmbH, Germany). Monocyte-derived DCswere used after 4 days of differentiation. DCs were incubatedfor 15 min with conditioned medium from the basolateral sideof CMT93/69 transwell cultures before addition of LPS (10 ng/ml,E. coli LPS, Sigma USA).

2.4. Electric cell–substrate impedance sensing (ECIS)

To measure intestinal epithelial barrier function, changes inelectrical resistance to alternating current flow were measured inreal time at 4 kHz using ECIS technology (Applied Biophysics, Troy,NY, USA). CMT93/69 cells were seeded on 8W10E+ electrode arrays,containing small gold-film electrodes, at a density of 105 cells perwell, and resistance measurements were directly started. After 48 h,when the epithelial cells had formed a barrier with a stable resis-tance of approximately 3500 �, T. suis E/S products were added. Tocalculate the contribution of paracellular resistance (mainly formedby tight junctions) to this total monolayer resistance, Rb valueswere calculated using ECIS software from Applied Biophysics. Theobtained graphs were normalized to the time point on which theT. suis E/S was added.

2.5. RNA isolation and real-time PCR

CMT93/69 cells were lysed in TRIzol. For RNA isolation phe-nol/chloroform extraction was performed using Phase Lock HeavyGel tubes (Eppendorf, Nijmegen, The Netherlands), followed byconsecutive isopropanol and ethanol precipitations. The concen-tration of isolated RNA, dissolved in RNAse free water, wasmeasured using a Nanodrop Spectrophotometer (Nanodrop Tech-nologies, Wilmington, DE, USA). An input of 1 �g RNA was usedfor cDNA synthesis using RevertAidTM First strand cDNA synthesiskit (Fermentas Life Sciences, Burlington, Canada), according to themanufacturer’s protocol. The resulting cDNA was used for analysis

by RT-PCR. mRNA expression levels were determined using an ABIprism 7900HT Sequence Detection System (PE Applied Biosystems,Foster city, CA, USA). Reactions were performed in a total reactionvolume of 10 �l, containing cDNA, 300 nM forward primer, 300 nM

I.H. Hiemstra et al. / Molecular Immunology 60 (2014) 1–7 3

Table 1Primer sequences.

Primer name Forward Reverse

ZO-1 GAA AGC AGA AGC CTC ATC TCC CAG GCT GTG ATG CGT GCGClaudin 3 TCA TCG GCA GCA GCA TCA TCA C CAG CAG CGA GTC GTA CAT TTT GCA CClaudin 4 CCT CTG GAT GAA CTG CGT GGT G CCC AGA GCA CCC ACG ATG ATGEMP1 GTCCTACGGCAATGAAGATGCT GAGGATCATGAAGGCTTGCACTTNF� TGG AAC TGG CAG AAG AGG CAC T CCA TAG AAC TGA TGA GAG GGA GGCCXCL1 ACCCAAACCGAAGTCATAGCC AGACAGGTGCCATCAGAGC

rtdasm

2

mgi

ssi

E4tRCd

2

tst

2

dc9o0s(

2

tgs*

Cyclo ACC CAT CAA ACC ATT CCT TCT GTA

everse primer, and SYBR Green PCR Mastermix (PE Applied Biosys-ems). Primers used for analysis of mRNA expression (Table 1) wereesigned using OligoExplorer1.2 software (www.GeneLink.com)nd synthesized by Invitrogen. Relative changes in mRNA expres-ion between samples were determined using the comparative CTethod (�CT).

.6. ELISA

Human TNF� protein levels in culture supernatants wereeasured using the CytoSetTM ELISA kit from Biosource (Invitro-

en, Bleiswijk, The Netherlands) according to the manufacturer’snstructions.

Murine CXCL1 protein levels in culture supernatants were mea-ured using the mouse CXCL1/KC Duoset ELISA kit from RnDystems (Abingdon, United Kingdom) according to manufacturer’snstructions.

To test effectiveness of the glycan oxidation procedure, T. suis/S (5 �g/ml in 0.2 M NaHCO3, pH 9.2) was coated overnight at◦C in a maxisorp 96-well plate (NUNC, Denmark). Coated pro-

eins were blocked using 1% Bovine Serum Albumin fraction V (BSA,oche Diagnostics, Germany). Binding of peroxidase-conjugatedoncanavalin A (ConA) lectin (5 �g/ml; EY Laboratories, USA) wasetected by TMB development.

.7. SDS-PAGE and silver staining

�-Chymotrypsin treated and untreated T. suis E/S (25 �g) pro-eins were separated by SDS-PAGE and subsequently visualized byilver staining of the gel. Silver staining was performed accordingo the manufacturers protocol (Bio-Rad, USA).

.8. Measurement of cell viability

Cell viability of CMT93/69 cells after T. suis E/S incubation wasetermined by an MTT assay. MTT (Sigma–Aldrich) was dissolved inell culture medium (0.5 mg/ml) and added to CMT93/69 cells. After0 min at 37 ◦C and 5% CO2, formed formazan crystals, as a resultf mitochondrial activity, were dissolved at room temperature in.1 M glycine/DMSO (6:1, pH 10.5). Absorption at 540 nm was mea-ured by photospectrometry using a Microplate reader Model 680Bio-Rad, Veenendaal, The Netherlands).

.9. Statistical evaluation

Values are expressed as the mean ± SEM. An ANOVA was used

o test for statistical significant differences between three or moreroups. A Student’s T-test was performed when comparing twoamples. A P-value of <0.05 was considered statistically significant.P < 0.05, **P < 0.01, and ***P < 0.001.

TGA GGA AAA TAT GGA ACC CAA AGA

3. Results

3.1. T. suis E/S products reduce intestinal epithelial barrierfunction

To investigate whether T. suis E/S components affect the bar-rier function of intestinal epithelial cells (IECs), monolayers of themurine IEC line CMT93/69 were incubated with T. suis E/S. Whenthe barrier function was measured using electric cell–substrateimpedance sensing (ECIS), a dose-dependent decrease in barrierfunction was observed upon incubation of the monolayers withT. suis E/S products (Fig. 1A). To determine whether the decreasein barrier function is accompanied by increased passage of solu-ble factors, IEC monolayers were apically incubated with T. suis E/Sproducts together with different sizes of FITC-labeled dextran. Pas-sage of the dextran through the IEC monolayers was determinedby measuring the fluorescence signal in the basolateral medium.Incubation of the monolayers with T. suis E/S products for 24 hincreased passage of different sizes of dextran through the mono-layers (Fig. 1B), compared to untreated monolayers. This suggeststhat these factors could actively pass the intestinal epitheliumto reach dendritic cells (DCs) in the underlying connective tis-sue of the intestines. In fact, soluble products of T. suis wormshave been shown to suppress the LPS-induced production of pro-inflammatory cytokines and chemokines by DCs (Klaver et al., 2013;Kuijk et al., 2012). To investigate whether these E/S compoundsare able to stimulate their own passage through the IEC mono-layer and are then still capable of causing the previously observedmodulation of DC function, basolateral medium from the epithelialtranswell cultures was harvested and added to human monocyte-derived DCs. After 15 min the DCs were stimulated with LPS for24 h, after which TNF� production was determined by ELISA. DCsthat had been incubated with basolateral medium from T. suisE/S-stimulated epithelial cells showed a significant reduction inLPS-induced TNF� levels, which was comparable to direct stim-ulation of DCs with 6 �g/ml T. suis E/S (Fig. 1C). Based on freeexchange of products between apical and basolateral sides, theapical addition of 50 �g/ml of T. suis E/S would result in a T. suisE/S concentration of 7.2 �g/ml in the basolateral medium, closelyresembling the suppression found in the DCs. This indicates that theepithelial barrier function is affected and that T. suis E/S may reachthe lamina propria and modulate DC functions. Alternatively, or inaddition, T. suis E/S products may induce the production of factorsby the IECs that suppress pro-inflammatory functions of DCs.

To exclude the possibility of cytotoxic effects of T. suis E/S onintestinal epithelial cells, cell viability was determined 24 h afterincubation of CMT93/69 monolayers with T. suis E/S products. Nodifference in cell viability was observed between cells that wereincubated with T. suis E/S products or unstimulated (Fig. 1D). Inaddition, nuclei of T. suis E/S-stimulated and unstimulated epithe-

lial monolayers were perfectly intact, as was shown by stainingwith the fluorescent nuclear dye DAPI (Fig. 1E), also indicating thatincubation with the helminth products does not induce IEC celldeath. These data indicate that T. suis E/S components can pass the

4 I.H. Hiemstra et al. / Molecular Immunology 60 (2014) 1–7

Fig. 1. Excreted/secreted T. suis products reduce IEC barrier function. (A) Barrier function of CMT93/69 monolayers in the presence of T. suis E/S was assessed using ElectricCell Substrate Impedance Sensing (ECIS). Graphs are normalized and shown from the time point of T. suis E/S addition. (B) Permeability of CMT93/69 monolayers wasdetermined in a transwell assay by the apical addition of FITC-labeled dextran of indicated sizes (1 mg/ml) in presence or absence of T. suis E/S products (50 �g/ml). After24 h the fluorescence signal was measured in the medium removed from the basolateral side. (C) CMT93/69 monolayers were incubated with T. suis E/S products (50 �g/ml)at the apical side in a transwell for 24 h. Subsequently, basolateral medium or T. suis E/S (in indicated concentrations), respectively, was added to human monocyte-derivedd e harvi ed usw sent 3

Im

3

tClmaep2ud

3g

ipdufwcgd(ttw

endritic cells. After 15 min LPS (10 ng/ml) was added. After 24 h supernatants werncubated for 24 h with T. suis E/S products (50 �g/ml). Cell viability was determin

ith T. suis E/S (50 �g/ml) and nuclei were stained using Dapi. Results shown repre

EC barrier, which would allow them to reach the lamina propria toodulate DC function.

.2. T. suis E/S products alter tight junctions

The barrier function of IECs is largely formed by tight junc-ions, which are protein complexes that close the paracellular cleft.laudin-3 and claudin-4 form tight junctions between IECs and are

inked to the actin cytoskeleton by zona occludens-1 (ZO-1). TheRNA expression levels of ZO-1 and claudin-3 were not changed

fter incubation with T. suis E/S products (Fig. 2A and B). Inter-stingly, mRNA expression of claudin-4 and epithelial membranerotein 1 (EMP-1), a claudin-like tight junction protein (Lee et al.,005; Bangsow et al., 2008), was significantly reduced after stim-lation with T. suis E/S components, which may contribute to theecreased epithelial barrier function (Fig. 2C and D).

.3. T. suis E/S mediated reduction in barrier function islycan-dependent

T. suis E/S products contain multiple complex molecules includ-ng many (glyco)proteins, which may function as proteases orore-forming proteins that could affect the epithelial integrity. Toetermine the role of T. suis E/S protein moieties in the down reg-lation of barrier function, T. suis E/S products were either heatedor 20 min at 80 ◦C to denature proteins (T. suis E/S DN), or treatedith �-chymotrypsin (T. suis E/S CT), which cleaves peptide chains

-terminal of aromatic amino acids. The involvement of T. suis E/Slycan moieties in the reduction of epithelial barrier function wasetermined by oxidation of these moieties using sodium periodate

T. suis E/S PI). Protein staining of T. suis E/S CT showed that pro-ein was largely broken down, indicating the effectiveness of thisreatment (Fig. 3D) Effectiveness of T. suis E/S periodate treatmentas shown by the loss of concanavalin A binding, which binds

ested and TNF� levels were determined by ELISA. (D) CMT93/69 monolayers wereing MTT solution (0.5 mg/ml). (E) CMT93/69 monolayers were stimulated for 24 h

separate experiments.

to oligomannose-type N-glycans (Fig. 3D). IEC monolayers wereincubated with these differently treated T. suis E/S products andbarrier function was followed in time by ECIS. Both DN- and CT-treated T. suis E/S products were still able to reduce the barrierfunction, excluding an essential involvement of protein moieties inthe reduction of epithelial barrier function (Fig. 3A and B). By con-trast, T. suis E/S PI no longer reduced IEC barrier function (Fig. 3C).Interestingly, periodate-sensitive T. suis glycans also appear to beessential in the suppression of EMP-1 and claudin-4, whereas DNor CT treatment of T. suis E/S products did not significantly altertheir effects (Fig. 3E and F). These data indicate that glycan compo-nents of T. suis E/S contribute to the reduction of epithelial barrierfunction.

3.4. Excreted/secreted T. suis products inhibit pro-inflammatorysignaling in intestinal epithelial cells

Since T. suis E/S products suppress the pro-inflammatoryresponse of DCs to LPS, we determined whether T. suis E/S productsmodulate the LPS response of IECs. In response to LPS stimulation,epithelial monolayers produced large amounts of TNF� and CXCL1,a chemoattractant for neutrophils (Mehrad et al., 1999). In con-trast, when the cells were pre-incubated with the E/S components,the production of the pro-inflammatory factors was significantlyreduced both at mRNA and protein levels (Fig. 4A–C). When incu-bated with denatured, �-chymotrypsin-, and periodate-treated T.suis E/S products, the observed effects were reversed, indicatingthat cytokine suppression depended on both protein and glycanmoieties.

To determine whether T. suis E/S signaling in IECs could lead

to the production of factors known to be induced by helminthinfection, we determined the expression of TSLP and Muc2 afterstimulation with T. suis E/S products. TSLP is an importantepithelium-derived factor for the stimulation of TH2 responses and

I.H. Hiemstra et al. / Molecular Immunology 60 (2014) 1–7 5

Fig. 2. Excreted/secreted T. suis products affect tight junction molecule expression. CMT93/69 monolayers were incubated with 50 �g/ml T. suis E/S for indicated time periods.m re detA te exp

i(alMT

LCs

Ft(adwte

RNA expression levels of ZO-1 (A), claudin-3 (B), claudin-4 (C) and EMP-1 (D) we. The mRNA levels of untreated CMT93/69 cells were set at 1.0. Results of 3 separa

s known to be induced in response to some helminth infectionsZaph et al., 2007). In addition, IECs are known to produce increasedmounts of mucus to induce expulsion of nematodes from the gutumen (Hasnain et al., 2010). mRNA expression of both TSLP and

uc2, the most abundant mucin in the colon, were induced upon. suis E/S stimulation (Fig. 4D and E).

In conclusion, these results show that T. suis E/S factors inhibit

PS-induced production of pro-inflammatory mediators TNF� andXCL1 by IECs, but stimulate the typical epithelial cell responsesuch as mucus and TSLP production.

ig. 3. T. suis-mediated reduction in barrier function is glycan-dependent. (A–C) Barrier fhe plateau phase of resistance, they were incubated with untreated T. suis E/S, denatuCT) T. suis E/S, all at 50 �g/ml. As a control, heat-inactivated chymotrypsin (CT) was incddition. (D) Proteins of �-chymotrypsin-treated and untreated T. suis E/S were separatedetermined by peroxidase-conjugated Concanavalin A (ConA) lectin (5 �g/ml) ELISA of peere incubated for 6 h with 50 �g/ml T. suis E/S, or DN, CT or PI treated E/S. mRNA expr

o expression levels of housekeeping gene cyclophilin A. The value for mRNA level of uxperiments.

ermined by RT-PCR relative to expression levels of housekeeping gene cyclophilineriments are shown.

4. Discussion

Previous studies have shown that soluble products of T. suismodulate DC function and suppress their capacity to induce TH1and TH17 immune responses (Kuijk et al., 2012). During infection,T. suis worms reside in the intestinal lumen and until now it wasunclear how these worms can affect DCs present in the lamina

propria. We here show that T. suis E/S products reduce the barrierfunction of the intestinal epithelial layer, thereby allowing passageof molecules over the cell barrier. This suggests that upon infection

unction of CMT93/69 monolayers was assessed using ECIS. After cells had reachedred (DN) T. suis E/S, periodate-treated (PI) T. suis E/S, or �-chymotrypsin-treatedluded. Graphs were normalized and shown starting from time point of T. suis E/S

by size on a SDS-PAGE gel and visualized by silver staining. Presence of glycans wasriodate-treated and untreated T. suis E/S (5 �g/ml). (E and F) CMT93/69 monolayersession levels of EMP-1 (E) and claudin-4 (F) were determined by RT-PCR relativentreated monolayers X is set at 1.0. Results shown represent data of 3 separate

6 I.H. Hiemstra et al. / Molecular Immunology 60 (2014) 1–7

Fig. 4. T. suis excreted/secreted proteins inhibit pro-inflammatory signaling in IECs. (A–C) CMT93/69 monolayers were incubated with untreated (U), denatured (DN), �-chymotrypsin-treated (CT) or periodate-treated (PI) T. suis E/S products. After 5 min LPS (100 ng/ml) was added. mRNA expression levels of TNF� (A) and CXCL1 (B) wered ilin A.E rationM gene

Eawat

rhpsclPswntbtott

sebooTi41ciit

etermined by RT-PCR relative to expression levels of housekeeping gene cyclophLISA (C). (D and E) CMT93/69 monolayers were incubated with indicated concentuc2 (E) were determined by RT-PCR relative to expression levels of housekeeping

/S products may rapidly enter the lamina propria, where they canffect immune cells and modulate immune responses. In addition,e show that T. suis E/S products suppress inflammatory cytokine

nd chemokine production in IECs in response to LPS and stimulatehe production of TSLP that enhances TH2 responses.

How T. suis E/S factors exactly reduce intestinal epithelial bar-ier function remains to be shown. It has been reported thatigh concentrations of T. suis E/S products reduced viability oforcine IECs when applied on both the apical and basolateralide (Abner et al., 2002). However, we did not observe signs ofell death when using 50 �g/ml of T. suis E/S products, a muchower concentration than used in other studies (Abner et al., 2002;arthasarathy and Mansfield, 2005). Also, no effects on the expres-ion and localization of tight junction proteins claudin-3 and ZO-1ere observed. Expression of claudin-4 and EMP-1 mRNA was sig-ificantly reduced after T. suis E/S incubation, suggesting that theseight junction proteins are involved in the reduction of epithelialarrier function. However, since the barrier function shows reduc-ion almost immediately after the start of stimulation, involvementf additional mechanisms such as conformational changes in theight junction structure or altered cellular localization of tight junc-ion proteins is plausible as well.

One mechanism that has been proposed for helminthic inva-ion in intestinal tissues is the ability of certain products that arexcreted or secreted by these helminths to form pores in lipidilayers. Indeed, from the 1288 proteins predicted to be excretedr secreted by T. suis 1.4% are possibly involved in the formationf pores in membranes (Cantacessi et al., 2011). Interestingly,richuris trichiura and Trichuris muris, the human and mouse-nfecting versions of T. suis, have been shown to excrete/secrete a7 kDa and 43 kDa pore-forming protein, respectively (Drake et al.,994). However, T. suis E/S in which proteins were denatured or

leaved by �-chymotrypsin could still reduce barrier function ofntestinal epithelial monolayers, indicating that T. suis proteins,ncluding pore-forming proteins, are not likely to be involved inhe observed reduction of intestinal epithelial barrier function.

CXCL1 protein expression levels in the culture supernatants were determined bys of T. suis E/S for indicated time periods. mRNA expression levels of TSLP (D) and

cyclophilin A. Results of 3 separate experiments are shown.

In contrast, oxidation of glycans of the T. suis E/S products couldabrogate this process, indicating that intact T. suis-derived glycansplay a role in the reduction of intestinal epithelial barrier functionthat was observed with untreated E/S products.

Helminths and their excreted/secreted factors are heavily gly-cosylated. This glycosylation comprises typical helminth glycansbut also host-like glycans. The latter are considered an importanttool for helminths in modulation of the host immune response byaffecting DC maturation. Presence of glycans has been shown tobe essential for helminths to induce TH2 responses (Tundup et al.,2012; van Die and Cummings, 2010). Helminth glycans can medi-ate their effects by binding host cell glycan binding proteins, suchas C-type lectin receptors or galectins. Whereas members of theselectin families are widely expressed by DCs and macrophages, theonly lectins known to be expressed in IECs are galectins (de Kivitet al., 2011) and intelectins (Tsuji et al., 2001, 2010). The mecha-nism by which glycans may be involved in the observed effects hereare not yet known, and subject for further studies.

T. suis soluble products have previously been shown to suppressthe LPS-induced production of TH1-inducing pro-inflammatorycytokines IL-12 and TNF� by DCs (Klaver et al., 2013; Kuijk et al.,2012). We here show a similar effect of T. suis E/S factors on LPS-induced TNF� and CXCL1 production by IECs. This suggests thathelminths suppress TH1 responses via their modulating effects onDCs as well as IECs. IECs may further contribute to the inductionof TH2 responses in the presence of T. suis E/S via the productionof TSLP. TSLP expression in IECs has previously been shown to beessential for inhibition of pro-inflammatory cytokine production byDCs and for the induction of TH2-responses during T. muris infection(Zaph et al., 2007). The increase of TSLP expression levels that weobserved upon T. suis E/S incubation was modest, however in vivomast cells have been shown to enhance the production of factors

like TSLP by IECs (Hepworth et al., 2012).

In Crohn’s disease patients, the epithelial barrier is disrupted byintestinal lesions as a result of inflammation (Silva, 2009) which isthought to contribute to pathology. Here, we unexpectedly show

ular Im

trmamacptrTi

bmcowoow

C

R

A

A

B

B

C

d

D

E

F

F

I.H. Hiemstra et al. / Molec

hat treatment with the helminth products T. suis E/S also leads to aeduction in IEC barrier function. However, we show that this treat-ent does not disintegrate the barrier, but instead allows ion fluxes

nd passage of (glyco)proteins via pore formation. In addition, theucus layer is known to play an essential role in the formation of

n intact epithelial barrier, by protecting the epithelial cells fromontact with bacteria. Helminths are known to stimulate mucusroduction (Wolff et al., 2012) and we here show that Muc2 produc-ion, a particularly prominent mucin within the gut, is induced inesponse to T. suis products. By strengthening the mucous layer, the. suis E/S products might compensate for a small loss in epithelialntegrity.

In conclusion, we show that T. suis E/S products reduce the IECarrier function thereby allowing passage of T. suis E/S immune-odulatory products that suppress LPS-induced pro-inflammatory

ytokine production by both IECs and DCs. Unraveling the identityf the T. suis E/S components that modulate IEC and DC functionill provide a useful tool to selectively skew immune responses by

rally administrated compounds and potentially reduce symptomsf autoimmune and allergic diseases without the need of infectionith live helminths.

onflict of interest

The authors declare no conflict of interest.

eferences

bner, S.R., Hill, D.E., Turner, J.R., Black, E.D., Bartlett, P., Urban, J.F., Mansfield, L.S.,2002. Response of intestinal epithelial cells to Trichuris suis excretory-secretoryproducts and the influence on Campylobacter jejuni invasion under in vitro con-ditions. J. Parasitol. 88, 738–745.

llen, J.E., Maizels, R.M., 2011. Diversity and dialogue in immunity to helminths. Nat.Rev. Immunol. 11, 375–388.

angsow, T., Baumann, E., Bangsow, C., Jaeger, M.H., Pelzer, B., Gruhn, P., Wolf, S.,von Melchner, H., Stanimirovic, D.B., 2008. The epithelial membrane protein 1is a novel tight junction protein of the blood–brain barrier. J. Cereb. Blood FlowMetab. 28, 1249–1260.

eer, R.J., 1976. The relationship between Trichuris trichiura (Linnaeus 1758) of manand Trichuris suis (Schrank 1788) of the pig. Res. Vet. Sci. 20, 47–54.

antacessi, C., Young, N.D., Nejsum, P., Jex, A.R., Campbell, B.E., Hall, R.S., Thams-borg, S.M., Scheerlinck, J.P., Gasser, R.B., 2011. The transcriptome of Trichurissuis—first molecular insights into a parasite with curative properties for keyimmune diseases of humans. PLoS ONE 6, e23590.

e Kivit, S., Kraneveld, A.D., Garssen, J., Willemsen, L.E., 2011. Glycan recognition atthe interface of the intestinal immune system: target for immune modulationvia dietary components. Eur. J. Pharmacol. 668 (Suppl. 1), S124–S132.

rake, L., Korchev, Y., Bashford, L., Djamgoz, M., Wakelin, D., Ashall, F., Bundy, D.,1994. The major secreted product of the whipworm, Trichuris, is a pore-formingprotein. Proc. Biol. Sci. 257, 255–261.

lliott, D.E., Li, J., Blum, A., Metwali, A., Qadir, K., Urban Jr., J.F., Weinstock, J.V., 2003.Exposure to schistosome eggs protects mice from TNBS-induced colitis. Am. J.Physiol. Gastrointest. Liver Physiol. 284, G385–G391.

eary, J., Britton, J., Leonardi-Bee, J., 2011. Atopy and current intestinal parasiteinfection: a systematic review and meta-analysis. Allergy 66, 569–578.

leming, J.O., Isaak, A., Lee, J.E., Luzzio, C.C., Carrithers, M.D., Cook, T.D., Field,A.S., Boland, J., Fabry, Z., 2011. Probiotic helminth administration in relapsing-remitting multiple sclerosis: a phase 1 study. Mult. Scler. 17, 743–754.

munology 60 (2014) 1–7 7

Flohr, C., Tuyen, L.N., Quinnell, R.J., Lewis, S., Minh, T.T., Campbell, J., Simmons, C.,Telford, G., Brown, A., Hien, T.T., Farrar, J., Williams, H., Pritchard, D.I., Britton,J., 2010. Reduced helminth burden increases allergen skin sensitization but notclinical allergy: a randomized, double-blind, placebo-controlled trial in Vietnam.Clin. Exp. Allergy 40, 131–142.

Franks, L.M., Hemmings, V.J., 1978. A cell line from an induced carcinoma of mouserectum. J. Pathol. 124, 35–38.

Hasnain, S.Z., Wang, H., Ghia, J.E., Haq, N., Deng, Y., Velcich, A., Grencis, R.K., Thornton,D.J., Khan, W.I., 2010. Mucin gene deficiency in mice impairs host resistance toan enteric parasitic infection. Gastroenterology 138, 1763–1771.

Hepworth, M.R., nilowicz-Luebert, E., Rausch, S., Metz, M., Klotz, C., Maurer, M.,Hartmann, S., 2012. Mast cells orchestrate type 2 immunity to helminthsthrough regulation of tissue-derived cytokines. Proc. Natl. Acad. Sci. U. S. A. 109,6644–6649.

Hewitson, J.P., Grainger, J.R., Maizels, R.M., 2009. Helminth immunoregulation: therole of parasite secreted proteins in modulating host immunity. Mol. Biochem.Parasitol. 167, 1–11.

Jouvin, M.H., Kinet, J.P., 2012. Trichuris suis ova: testing a helminth-based therapy asan extension of the hygiene hypothesis. J. Allergy Clin. Immunol. 130, 3–10.

Klaver, E.J., Kuijk, L.M., Laan, L.C., Kringel, H., van Vliet, S.J., Bouma, G., Cummings,R.D., Kraal, G., van Die, I., 2013. Trichuris suis-induced modulation of humandendritic cell function is glycan-mediated. Int. J. Parasitol. 43, 191–200.

Kuijk, L.M., Klaver, E.J., Kooij, G., van der Pol, S.M., Heijnen, P., Bruijns, S.C., Kringel,H., Pinelli, E., Kraal, G., de Vries, H.E., Dijkstra, C.D., Bouma, G., van Die, I.,2012. Soluble helminth products suppress clinical signs in murine experimentalautoimmune encephalomyelitis and differentially modulate human dendriticcell activation. Mol. Immunol. 51, 210–218.

Lee, H.S., Sherley, J.L., Chen, J.J., Chiu, C.C., Chiou, L.L., Liang, J.D., Yang, P.C., Huang,G.T., Sheu, J.C., 2005. EMP-1 is a junctional protein in a liver stem cell line andin the liver. Biochem. Biophys. Res. Commun. 334, 996–1003.

Mehrad, B., Strieter, R.M., Moore, T.A., Tsai, W.C., Lira, S.A., Standiford, T.J., 1999.CXC chemokine receptor-2 ligands are necessary components of neutrophil-mediated host defense in invasive pulmonary aspergillosis. J. Immunol. 163,6086–6094.

Parthasarathy, G., Mansfield, L.S., 2005. Trichuris suis excretory secretory products(ESP) elicit interleukin-6 (IL-6) and IL-10 secretion from intestinal epithelial cells(IPEC-1). Vet. Parasitol. 131, 317–324.

Pearce, E.J., Kane, M., Sun, J., Taylor, J., McKee, A.S., Cervi, L., 2004. TH2 responsepolarization during infection with the helminth parasite Schistosoma mansoni.Immunol. Rev. 201, 117–126.

Silva, M.A., 2009. Intestinal dendritic cells and epithelial barrier dysfunction inCrohn’s disease. Inflamm. Bowel Dis. 15, 436–453.

Summers, R.W., Elliott, D.E., Urban Jr., J.F., Thompson, R., Weinstock, J.V., 2005a.Trichuris suis therapy in Crohn’s disease. Gut 54, 87–90.

Summers, R.W., Elliott, D.E., Urban Jr., J.F., Thompson, R.A., Weinstock, J.V., 2005b.Trichuris suis therapy for active ulcerative colitis: a randomized controlled trial.Gastroenterology 128, 825–832.

Tsuji, S., Uehori, J., Matsumoto, M., Suzuki, Y., Matsuhisa, A., Toyoshima, K., Seya, T.,2001. Human intelectin is a novel soluble lectin that recognizes galactofuranosein carbohydrate chains of bacterial cell wall. J. Biol. Chem. 276, 23456–23463.

Tsuji, S., Tsuura, Y., Morohoshi, T., Shinohara, T., Oshita, F., Yamada, K., Kameda, Y.,Ohtsu, T., Nakamura, Y., Miyagi, Y., 2010. Secretion of intelectin-1 from malig-nant pleural mesothelioma into pleural effusion. Br. J. Cancer 103, 517–523.

Tundup, S., Srivastava, L., Harn Jr., D.A., 2012. Polarization of host immune responsesby helminth-expressed glycans. Ann. N. Y. Acad. Sci. 1253, E1–E13.

van Die, I., Cummings, R.D., 2010. Glycan gimmickry by parasitic helminths: a strat-egy for modulating the host immune response? Glycobiology 20, 2–12.

van Stijn, C.M., van den Broek, M., van de Weerd, R., Visser, M., Tasdelen, I., Tefsen, B.,van Die, I., 2009. Regulation of expression and secretion of galectin-3 in humanmonocyte-derived dendritic cells. Mol. Immunol. 46, 3292–3299.

Wolff, M.J., Broadhurst, M.J., Loke, P., 2012. Helminthic therapy: improving mucosal

barrier function. Trends Parasitol. 28, 187–194.

Zaph, C., Troy, A.E., Taylor, B.C., Berman-Booty, L.D., Guild, K.J., Du, Y., Yost,E.A., Gruber, A.D., May, M.J., Greten, F.R., Eckmann, L., Karin, M., Artis, D.,2007. Epithelial-cell-intrinsic IKK-beta expression regulates intestinal immunehomeostasis. Nature 446, 552–556.