Increased intraepithelial V a 24 Invariant NKT cells in the celiac duodenum

Article Increased Intraepithelial Vα24 Invariant NKT Cells in the Celiac Duodenum Enrique Montalvillo 1, David Bernardo 2,3, Beatriz Martínez-Abad 1, Yessica Allegretti 4, Luis Fernández-Salazar 5, Carmen Calvo 6, Fernando G. Chirdo 4, José A. Garrote 1,7,:and Eduardo Arranz 1,:,* Received: 23 July 2015 ; Accepted: 21 October 2015 ; Published: 30 October 2015 1Mucosal Immunology Lab, IBGM, University of Valladolid-CSIC, Sanz y Forés 3, 47003 Valladolid, Spain; [email protected] (E.M.); [email protected] (B.M.A.); [email protected] (J.A.G.) 2Antigen Presentation Research Group, Imperial College London, Northwick Park & St Mark’s Campus, Level 7W, St Mark's Building Watford Road Harrow HA1 3UJ, UK; d.bernar[email protected] 3Gastroenterology Unit, Hospital Universitario de La Princesa and Instituto de Investigación Sanitaria Princesa (IIS-IP), Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBEREHD), Madrid 28006, Spain 4Laboratorio de Investigación en el Sistema Inmune –LISIN, Departamento de Ciencias Biológicas, Facultad de Ciencias Exactas, Universidad Nacional de La Plata, 115, La Plata 1900, Buenos Aires, Argentina; [email protected] (Y.A.); [email protected] (F.G.C.) 5Gastroenterology Unit, Hospital Clínico Universitario; Ramón y Cajal 3, Valladolid 47005, Spain; [email protected] 6Paediatric Unit, Hospital Clínico Universitario; Ramón y Cajal 3, Valladolid 47005, Spain; carmencalvoromer[email protected] 7Medical Laboratory Service, Hospital Universitario Rio Hortega; Dulzaina 2, Valladolid 47012, Spain *Correspondence: [email protected]; Tel.: +34-983-184-843 :These authors contributed equally to this work. Abstract: Celiac Disease (CD) is an interferon (IFN)γ-mediated duodenal hypersensitivity to wheat gluten occurring in genetically predisposed individuals. Gluten-free diet (GFD) leads to a complete remission of the disease. Vα24-restricted invariant NKT (iNKT) cells are important to maintain immune homeostasis in the gut mucosa because of their unique capacity to rapidly produce large quantities of both T-helper (Th)1 and Th2 cytokines upon stimulation. We studied the presence of these cells in the CD duodenum. Duodenal biopsies were obtained from 45 untreated-CD patients (uCD), 15 Gluten Free Diet-CD patients (GFD-CD), 44 non-inflamed non-CD controls (C-controls) and 15 inflamed non-CD controls (I-controls). Two populations from Spain and Argentina were recruited. Messenger RNA (mRNA) expression of Vα24-Jα18 (invariant TCRαchain of human iNKT cells), IFNγand intracellular transcription factor Forkhead Box P3 (Foxp3), and flow cytometry intraepithelial lymphocyte (IEL) profile were determined. Both uCD and GFD-CD patients had higher Vα24-Jα18 mRNA levels than non-CD controls (I and C-controls). The expression of Vα24-Jα18 correlated with Marsh score for the severity of mucosal lesion and also with increased mRNA IFNγlevels. uCD and GFD-CD patients had decreased mRNA expression of FoxP3 but increased expression of Vα24-Jα18, which revealed a CD-like molecular profile. Increased numbers of iNKT cells were confirmed by flow cytometry within the intraepithelial lymphocyte compartment of uCD and GFD-CD patients and correlated with Vα24-Jα18 mRNA expression. In conclusion, we have found an increased number of iNKT cells in the duodenum from both uCD and GFD-CD patients, irrespective of the mucosal status. A CD-like molecular profile, defined by an increased mRNA expression of Vα24-Jα18 together with a decreased expression of FoxP3, may represent a pro-inflammatory signature of the CD duodenum. Nutrients 2015,7, 8960–8976; doi:10.3390/nu7115444 www.mdpi.com/journal/nutrients Nutrients 2015,7, 8960–8976 Keywords: Celiac Disease; Intraepithelial Lymphocytes; iNKT; Vα24-Jα18; IFNγ; Celiac Disease-like molecular profile 1. Introduction Celiac disease (CD) is an inflammatory disorder of the small intestine induced by wheat gluten and other prolamins from rye, barley and some varieties of oats [1] in genetically susceptible individuals. It is characterized by an interferon (IFN)-γmediated type I cytokine profile [2]. CD manifestation is characterized by an increased number of intraepithelial and lamina propria lymphocytes, villous atrophy, tissue remodeling and the presence of anti-transglutaminase antibodies [3]. At present, the only treatment for CD is a life-long strict gluten-free diet (GFD), which normally leads to a complete remission of the disease [4]. Gut intraepithelial lymphocytes (IEL) comprise a heterogeneous population of cells outside the normal circulation. In addition to conventional T lymphocytes (CD3`TCRαβ`, either CD4`or CD8`) with an outstanding CD3`CD8`prevalence, Natural Killer (NK) cells and unconventional IEL populations such as CD8αα, TCRγδ`cells, CD3`CD4´CD8´cells and NKT lymphocytes are widely represented [5,6]. Increased total numbers of IEL CD3`(both classical and TCRγδ`) and decreased IEL non-T cells (CD3´, CD103`) have been consistently reported in CD [7–9]. A direct cytolytic effect of conventional CD3`αβ`CD8`cytotoxic IEL on adjacent enterocytes is undisputed [10,11] and related to villous atrophy [7]. However, the role of IEL TCRγδ`in CD pathogenesis remains elusive, although they may have a key role in oral tolerance as supported by the identification of a subset of regulatory TCRγδ`IEL population capable of limiting the cytotoxicity of IEL in CD treated on a GFD [12]. Classical regulatory Tcells (Tregs; CD3`, CD4`, CD25`and intracellular transcription factor Forkhead Box P3`(FoxP3`)) and non-classical interleukin (IL)-10 producing regulatory cells (Tr1: CD3`, CD4`, CD25´and intracellular FoxP3´) are the main regulatory T-cells found in the intestine [13]. Tregs elicit their function by suppressing IL-2 production and T-cell proliferation [14], while Tr1 cells are the main source of IL-10 in the intestinal lamina propria since they are chronically stimulated and limit the production of pro-inflammatory cytokine by controlling inflammatory responses to dietary antigens. Compared to Tregs, the finding of larger numbers of Tr1 in the intestinal lamina propria suggests that these cells have an important regulatory capacity [15,16]. Invariant NKT cells (iNKT; CD3`, TCR Vα24`Vβ11`) are also important to maintain immune homeostasis [13]. Human iNKT cells express classical NK cell markers as well as an invariant TCRα chain (iNKTα) (Vα24-Jα18 in humans) paired to “semi-invariant” TCRβchains (iNKTβ), which recognizes antigens presented by the major histocompatibility complex (MHC) class I-like molecule CD1d [17,18]. For all iNKT-cell TCRs, binding to CD1d is primarily mediated by the Vα-Jαrearranged invariant CDR3αloop [19]. Therefore, the anti-Vα24-Jα18 is the standard method used to detect human iNKT cells [20,21]. These cells can be sub-divided into CD4`and CD4´(most of these CD4´CD8´) cells. CD4´CD8´iNKT cells produce predominantly T-helper (Th)1 cytokines (IFNγ and TNFα) whereas CD4`iNKT cells can produce both Th1 and Th2 (IL-4 and IL-13) cytokines [13]. Because of their unique capacity to rapidly produce large quantities of both Th1 (IFNγ) and Th2 (IL-4) cytokines upon stimulation [22], iNKT cells may have a key role in protection against tumors or in preventing autoimmune disease [23]. Despite low numbers, iNKT cells have a central role in intestinal homeostasis [17,24,25] and are essential for the development of oral tolerance [26,27]. Nevertheless, their number within the intraepithelial and lamina propria compartments and their specific role in CD pathogenesis remains elusive. In this manuscript, we aimed to study whether changes in the number of iNKT cells may be altered in the duodenum of CD patients. To these aim we assessed the mRNA expression of 8961 Nutrients 2015,7, 8960–8976 Vα24-Jα18 and the proportion of iNKT cells within the intraepithelial compartment to reveal an increased number of these cells in the CD mucosa. 2. Materials and Methods 2.1. Patients and Biopsy Samples Duodenal samples were collected from two independent populations in Spain (Hospital Clínico Universitario de Valladolid) and Argentina (Biobank from the LISIN, La Plata). The Spanish population included 25 untreated celiac patients (uCD, mean age 28.9 years; range 5–76 years; 42% males) (Table S1), 15 CD patients treated with GFD (GFD-CD; mean age 34.2 years; range 4-71 years; 34% males) (Table S2), 15 non-CD patients with other inflamed conditions (I-controls, mean age 42.1 years; range 15–78 years; 56% males) (Table S3) and 25 non-inflamed non-CD controls (C-controls; mean age 38.3 years; range 6–81 years; 30% males) (Table S4). The Argentinian population included 20 uCD patients (mean age 24.8 years; range 4-56 years; 28% males) (Table S5) and 19 C-controls (mean age 31.4 years; range 6–62 years; 52% males) (Table S6). Regarding age and gender, no statistically differences were found between Spanish and Argentinian patients. Clinical data from patient groups included in the study are shown in Table 1. The experiments were conducted with the understanding and the written consent of the adult participants, or the next of kin, caretakers, or guardians on behalf of the minors/children enrolled in this study. The study and the written consent procedure were approved by the Ethics committees from Hospital Clínico Universitario of Valladolid and Biobank from the LISIN, La Plata. Table 1. Clinical data from patient groups included in the study. Study Patients nMean age (Range) Gender HLA DQ2/DQ8 IgA antitTG/EMA Marsh Criteria at Diagnosis GFD Celiac Patients uCD 45 27.1 (4–76) 36% males + + II-III No GFD-CD 15 34.2 (4–71) 34% males + - II-III Yes Non Celiac Patients I-controls 15 42.1 (15–78) 56% males +/- - (*A)0-I (M.M) No C-controls 44 35.3 (6–81) 41% males - (*B) - (*A)0 No uCD (untreated celiac patients), GFD-CD (celiac patients treated with gluten free diet), I-controls (non celiac patients with other inflamed conditions) and C-controls (non-celiac disease patients without other inflamed conditions). IgA anti-tTG (Anti-tissue transglutaminase antibodies), EMA (Endomysium antibodies), GFD (Gluten free diet). M.M (mild mucosal alterations non compatibles with celiac disease). *ASerological test were performed only in genetically susceptible patients. *BTwo patients had positive genetic susceptibility markers. At diagnosis, all CD patients had CD-compatible symptoms, positive anti-endomysium and/or anti-transglutaminase IgA antibodies, CD-associated risk alleles (HLA-DQ2 and DQ8), and duodenal biopsy with histopathological changes. No differences in clinical markers were found between Spanish and Argentinian CD individuals. Patients on a GFD showed an improvement of the histological lesion (Marsh 0-I), and negative serum anti-transglutaminase antibodies for at least one year. Control groups were collected from patients referred to the gastroenterology clinics for diagnostic investigations due to clinical suspicion of intestinal disease (chronic diarrhea, gastritis by Helicobacter pylori, hiatus hernia, etc.). Similar symptoms were observed in both populations. Some of these cases showed duodenal inflammation (I-controls) while lack of mucosal affection was found in C-controls. None of them had a final diagnosis of CD. 2.2. Quantitative PCR Duodenal biopsies from the Spanish (40 CD patients (25 uCD and 15 GFD-CD), 25 C-controls and 15 I-controls) and the Argentinian population (20 uCD and 19 C-controls) were submerged in 0.5 mL of RNALaterrsolution (Ambion Inc, Austin, Texas, USA) and stored at ´20 ˝C immediately after sample taking. Total RNA was isolated using the TRI-ReagentrSolution according to manufacturer 8962 Nutrients 2015,7, 8960–8976 instructions (Ambion Inc, Austin, Texas, USA). In parallel, 15 duodenal samples from each group of patients from the Spanish population were also analyzed by flow cytometry as described below to determine the phenotype of lymphocytes and iNKT cells. Reverse transcription was carried out by using the SuperScriptrFirst-Strand Synthesis System for reverse transcriptase (RT)-PCR Kit (Applied Biosystems, Carlsbad, CA, USA) with random hexamers as primers. Reactions were performed using the FastStart SYBR Green MasterMix (Roche Applied Science, Mannheim, Germany) with thermolabile Uracil DNA Glycosylase to prevent carry-over contamination. Messenger RNA levels (βactin, IFNγ, Vα24-Jα18 and FoxP3) were measured by quantitative PCR (qPCR) on a LightCyclerrinstrument (Roche Applied Science, Mannheim, Germany) after extrapolation to an external curve. Primer sets and PCR conditions are described in Table 2. Levels of mRNA are expressed as the ratio molecule/βactin in arbitrary units (AU). Table 2. Primer sequences for quantitative-PCR. Molecule Primers Sequence NCBI Locus Annealing T βactin fw: 51ATG GGT CAG AAG GAT TCC TAT GTG - 31 rv: 51CTT CAT GAG GTA GTC AGT CAG GTC - 31NM_001101.3 60 IFNγfw: 51TGG AAA GAG GAG AGT GAC AG - 31 rv: 51ATT CAT GTC TTC CTT GAT GG - 31NM_000619.2 60 Vα24-Jα18 fw: 51CTG GAG GGA AAG AAC TGC - 31 rv: 51TGT CAG GGA AAC AGG ACC - 31NC_000014.9 65 FoxP3 fw: 51CAG CAC ATT CCC AGA GTT CCT C - 31 rv: 51GCG TGT GAA CCA GTG GTA GAT C - 31NM_014009.3 60 Primer sequences used for quantitative-PCR. NCBI locus and annealing temperature point (annealing T). IFNγ (interferon-γ), Vα24-Jα18 (invariant TCRαchain of human iNKT cells), FoxP3 (intracellular transcription factor Forkhead Box P3). 2.3. Isolation of Intraepithelial Lymphocytes and Lamina Propria Mononuclear Cells Biopsy samples from 15 uCD, 15 GFD, 15 non-inflamed non-CD controls (C-controls) and 15 inflamed non-CD controls (I-controls) were collected from the Spanish population. Samples were kept in ice-chilled physiologic phosphate buffered saline (PBS) and processed within an hour as previously described [28,29]. Briefly, IEL and epithelial cells were released from the mucosal specimens by incubation for 1 hour under gentle agitation with 1 mM ethylenediaminetetraacetic acid (EDTA) and 1 mM dithiothreitol (DTT) in RPMI 1640 medium (GibcoBRL Life Technologies, Vienna, Austria) supplemented with 10% fetal calf serum, 2mM L-glutamine, 100U/mL penicillin, 100 µg/mL streptomycin and 0.25 µg/mL amphotericin (GibcoBRL Life Technologies, Vienna, Austria). Following DTT and EDTA incubation IEL were released into the medium and collected by centrifugation, washed twice in PBS (Lonza, Braine-l’Alleud, Belgium) and stained with fluorochrome-conjugated monoclonal antibodies (mAbs). The remaining tissue was incubated in moderated rotation at 37 ˝C for 90–120 min. with 1 mg/mL of collagenase D in RPMI 1640 medium (GibcoBRL Life Technologies, Vienna, Austria) supplemented with 10% fetal calf serum and antibiotics until the biopsies have been completely degraded. Single cell suspensions were filtered (70 µm Nylon Filter, BD Biosciences, San Diego, CA, USA) to remove non-cellular fibers, and the lamina propria mononuclear cells (LPMC) suspension was washed twice in PBS. 2.4. Antibody Labeling and Flow Cytometry Analysis A total of 100,000 isolated cells (IEL or LPMC) were labeled with fluorochrome-conjugated monoclonal antibodies (mAbs) and their appropriate isotype-matched control antibodies from the same manufacturers. The fluorochrome-conjugated mAbs were: FITC Mouse anti-human CD103 (clone Ber-ACT8), PE Mouse anti-human Vα24-Jα18 (clone 6B11, specifically recognizing all T cells expressing the conserved CDR3 region of the Vα24Jα18 invariant TCRαrearrangement), PE 8963 Nutrients 2015,7, 8960–8976 Mouse anti-human TCRγδ (clone B1), APC Mouse anti-human CD3 (clone HIT3a), APC Mouse anti-human CD25 (clone M-A251), PE-Cy7 Mouse anti-human CD8 (clone RPA-T8) and PE-Cy7 Mouse anti-human CD45 (clone HI30) from BD Pharmingen (San Diego, CA, USA); PE Mouse anti-human FoxP3 (clone PCH101) from eBioscience (San Diego, CA, USA); FITC Mouse anti-human CD4 (clone 13B8.2) and PE Mouse anti-human CD8 (clone B9.11) from Beckman Coulter (Brea , CA, USA). Cells were labeled in phosphate-buffered saline containing 1 mM EDTA and 0.02% sodium azide (fluorescent-activated cell sorting (FACS) buffer). Labeling was performed on ice and in the dark for 20 min. Cells were washed twice in FACS buffer, fixed with 1% paraformaldehyde in 0.85% saline, and stored at 4 ˝C before acquisition on the flow cytometer within 24 h. For FoxP3 intracellular staining, cells were fixed with Leucoperm A following surface staining and permeabilized with Leucoperm B (Bio-Rad, UK) before adding antibody for intracellular labeling. After incubation cells were washed in FACS buffer, fixed, and acquired as previously reported. Cells were acquired in a Beckman Coulter FC500 flow cytometer and data processed with Cell BC software (Beckman Coulter, Brea, CA, USA). All IEL and lamina propria lymphocyte (LPL) cells were identified as CD45`(leukocyte pan-marker) and IELs were also identified as CD103`. Non-T cells (CD3´), TCRγδ cells (CD3`TCRγδ`), TCRαβ cells (CD3`TCRγδ´) (Figure 1A), iNKT cells (CD3`Vα24-Jα18`) (Figure 1B) and Treg cells (CD3`CD4`CD25`FoxP3`or CD3`CD4`FoxP3`) were identified by flow cytometry within the intraepithelial and the lamina propria compartments. Numbers of cells were expressed as percentages. Nutrients 2015,7,page–page 5 CD25(cloneM‐A251),PE‐Cy7Mouseanti‐humanCD8(cloneRPA‐T8)andPE‐Cy7Mouse anti‐humanCD45(cloneHI30)fromBDPharmingen(SanDiego,CA,USA);PEMouseanti‐human FoxP3(clonePCH101)fromeBioscience(SanDiego,CA,USA);FITCMouseanti‐humanCD4(clone 13B8.2)andPEMouseanti‐humanCD8(cloneB9.11)fromBeckmanCoulter(Brea,CA,USA).Cells werelabeledinphosphate‐bufferedsalinecontaining1mMEDTAand0.02%sodiumazide (fluorescent‐activatedcellsorting(FACS)buffer).Labelingwasperformedoniceandinthedarkfor 20min.CellswerewashedtwiceinFACSbuffer,fixedwith1%paraformaldehydein0.85%saline, andstoredat4°Cbeforeacquisitionontheflowcytometerwithin24h.ForFoxP3intracellular staining,cellswerefixedwithLeucopermAfollowingsurfacestainingandpermeabilizedwith LeucopermB(Bio‐Rad,UK)beforeaddingantibodyforintracellularlabeling.Afterincubationcells werewashedinFACSbuffer,fixed,andacquiredaspreviouslyreported. CellswereacquiredinaBeckmanCoulterFC500flowcytometeranddataprocessedwithCell BCsoftware(BeckmanCoulter,Brea,CA,USA).AllIELandlaminaproprialymphocyte(LPL)cells wereidentifiedasCD45 + (leukocytepan‐marker)andIELswerealsoidentifiedasCD103 + .Non‐T cells(CD3 − ),TCRγδ cells(CD3 + TCRγδ + ),TCRαβ cells(CD3 + TCRγδ − )(Figure1A),iNKTcells (CD3 + Vα24‐Jα18 + )(Figure1B)andTregcells(CD3 + CD4 + CD25 + FoxP3 + orCD3 + CD4 + FoxP3 + )were identifiedbyflowcytometrywithintheintraepithelialandthelaminapropriacompartments. Numbersofcellswereexpressedaspercentages.  Figure1.Identificationofintraepithelialandlaminaproprialymphocytesbyflowcytometry. Exampleofcharacterizationofintraepitheliallymphocytes(IELs)/laminaproprialymphocytes(LPL) inanuntreatedceliacdiseasedonor.Laminapropria(CD45 + )andIEL(CD45 + CD103 + )wereidentified andpercentagesofTCRαβ cells,TCRγδ cellsandnon‐Tcellsdetermined(A).Exampleof characterizationofinvariantNKT(iNKT)cellsinanuntreatedceliacdiseasedonor:CD45 + CD3 + Vα24‐ Jα18 + cellswithinthetotalofCD45 + CD3 + cells,iNKT(CD45 + CD3 + Vα24‐Jα18 + )phenotypeaccordingto theexpressionofCD4and/orCD8withinthetotalofiNKTs(B).  Figure 1. Identification of intraepithelial and lamina propria lymphocytes by flow cytometry. Example of characterization of intraepithelial lymphocytes (IELs)/lamina propria lymphocytes (LPL) in an untreated celiac disease donor. Lamina propria (CD45`) and IEL (CD45`CD103`) were identified and percentages of TCRαβ cells, TCRγδ cells and non-T cells determined (A). Example of characterization of invariant NKT (iNKT) cells in an untreated celiac disease donor: CD45`CD3`Vα24-Jα18`cells within the total of CD45`CD3`cells, iNKT (CD45`CD3`Vα24-Jα18`) phenotype according to the expression of CD4 and/or CD8 within the total of iNKTs (B). 2.5. Statistical Analysis Correlation analyses and two-tailed non-parametric statistical analyses were performed using the Kruskal-Wallis one way analysis of variance test, the Mann-Whitney U test and the non-parametric Spearman’s correlation. p< 0.05 was considered significant. Flow cytometry results were expressed as percentages and analysed by the two-tailed non-parametric Mann-Whitney U test. p< 0.05 was considered significant. Reference values of the IEL subpopulations for each patient’s group were expressed as median percentages with the interquartile range (IQR). 8964 Nutrients 2015,7, 8960–8976 3. Results 3.1. Increased Duodenal Vα24-Jα18 mRNA Expression in Celiac Disease Patients Due to the low numbers of iNKT cells both in blood [21] and tissue samples [22] we first studied the proportion of iNKT cells in duodenal biopsies by assessing the mRNA expression or their restrictive invariant Vα24-Jα18 chain in complete biopsy explants [20,21]. Untreated-CD patients (uCD) had increased Vα24-Jα18 mRNA levels compared to both inflamed non-CD (I-controls) (p < 0.05) and non-inflamed non-CD controls (C-controls) (p < 0.001) (Figure 2A). Gluten Free Diet-CD patients (GFD-CD) had increased Vα24-Jα18 mRNA levels compared to C-controls (p < 0.001). No differences were found between uCD and GFD-CD patients, which suggests an increased load of iNKT cells in duodenal biopsies from both groups of patients irrespectively of the disease status (Figure 2A). We also analyzed duodenal expression of Vα24-Jα18 mRNA in an independent population from Argentina. Figure 2B confirms that CD patients show increased duodenal Vα24-Jα18 mRNA levels compared to non-inflamed non-CD controls (p < 0.001), irrespectively of the origin of the samples. No statistically significant differences in Vα24-Jα18 mRNA levels were found between the Spanish and the Argentinian populations, neither within the control nor the untreated CD patient groups. Samples from GFD-CD patients showed a Marsh score for the severity of the mucosal lesion between 0 and I, and no differences were found in these sub-groups regarding Vα24-Jα18 mRNA expression. Untreated CD patients had a Marsh score rating from I to IIIc, and in these patients, Vα24-Jα18 mRNA levels correlated with the Marsh score (Spearman’s r= 0.063, p < 0.05) (Figure 2C). 3.2. Correlation between Duodenal Vα24-Jα18 and IFNγmRNA Expression in Celiac Disease Duodenal samples from CD patients (either treated and untreated) had increased IFNγmRNA expression compared with non-inflamed non-CD controls both in the Spanish (GFD-CD, p < 0.05; uCD, p < 0.001) (Figure 3A) and the Argentinian populations (uCD, p < 0.05) (Figure 3B) as previously described [30,31]. Since both mRNA levels of IFNγ[2,30] and Vα24-Jα18 (Figure 2C) correlated with the Marsh score for the severity of the mucosal lesion, we studied if the expression of IFNγand Vα24-Jα18 was related. A correlation was found between the mRNA levels of IFNγ and Vα24-Jα18 in treated and untreated CD samples in both the Spanish (C-controls: Spearman’s r= 0.3115, pvalue = n.s; I-controls: Spearman’s r= 0.4265, pvalue = n.s; GFD-CD: Spearman’s r= 0.5393, p < 0.05; uCD: Spearman’s r= 0.4323, p < 0.05) and the Argentinian populations (C-controls: Spearman’s r= 0.5895, p < 0.05; uCD: Spearman’s r= 0.6917, p < 0.001) (Figure 3C,D). 3.3. Duodenal Vα24-Jα18 and FoxP3 mRNA Levels Reveal a Celiac Disease Molecular Profile We next studied FoxP3 mRNA expression in the duodenum as an indirect way of quantifying Treg cells in tissue. CD samples (both uCD and GFD-CD) had decreased FoxP3 expression compared with non-inflamed non-CD-controls (C-controls) (uCD, p < 0.001; GFD-CD, p < 0.001) (Figure 4A). Similar results were found on a second independent analysis of the Argentinian population (uCD, p < 0.001) (Figure 4B). Because CD samples were characterized by increased duodenal mRNA expression of Vα24-Jα18 (Figure 2A,B) and decreased expression of FoxP3 (Figure 4A,B), we studied whether the joint analysis of these two molecules could help us to identify a CD-like molecular profile. No differences were found between samples from controls or uCD patients in any of the molecules studied in both populations (Spanish and Argentinian). Similar results were also obtained when Vα24-Jα18 and FoxP3 duodenal mRNA expression were independently analysed. Therefore, both populations were merged to increase the sample size of uCD and C-control groups. The joint analysis of both duodenal FoxP3 and Vα24-Jα18 mRNA levels allowed us to discriminate between CD (treated and untreated), and non-CD samples (inflamed or non-inflamed) with a sensibility and sensitivity of 92%, revealing a CD-like molecular profile (C-controls: Spearman’s r= 0.3495, p < 0.05; 8965 Nutrients 2015,7, 8960–8976 I-controls: Spearman’s r=´0.4577, pvalue = n.s; GFD-CD: Spearman’s r=´0.4858, p < 0.05; uCD: Spearman’s r=´0.056, pvalue = n.s) (Figure 4C). Nutrients 2015,7,page–page 7  Figure2.IncreasedduodenalVα24‐Jα18mRNAexpressioninuntreatedandtreatedceliacdisease patients.ExpressionlevelsofVα24‐Jα18mRNAinduodenalbiopsiesindicatedbytheratio Vα24‐Jα18/βactininarbitraryunits(AU),inuntreatedCeliacDisease(uCD),GlutenFreeDiet‐CD patients(GFD‐CD),inflamednon‐CDcontrols(I‐controls)andnon‐inflamednon‐CDcontrols (C‐controls)intheSpanishpopulation(A)andinuCDandC‐controlsintheArgentinianpopulation (B).Statisticallysignificantdifferencesareshown(twotailedMann‐WhitneyUtest;Krustall‐Wallis test).Horizontalbarsaremedianvalues.Correlationbetweenthedegreeofhistologicallesion(Marsh score)andtheexpressionlevelofV24‐Jα18mRNAinCDpatients.(1:Marsh0‐I,2:MarshII,3:Marsh IIIa,4:MarshIIIb,5:MarshIIIc)TherewasanincreasedofVα24‐Jα18expressionandtheduodenal increasedlevelofatrophy(Spearmanr=0.063,p<0.05).GFD‐CDpatientsshowaMarsh0‐I,uCD showduodenalatrophyMarshIItoIIIc(C). Figure3.CorrelationbetweenduodenalmRNAexpressionofVα24‐Jα18andIFNγ.Expressionlevels ofIFNγmRNAinduodenalbiopsiesindicatedastheratioIFNγ/βactin,inarbitraryunits(AU),in untreatedCeliacDisease(uCD),GlutenFreeDiet‐CDpatients(GFD‐CD),inflamednon‐CDcontrols (I‐controls)andnon‐inflamednon‐CDcontrols(C‐controls)intheSpanishpopulation(A)andinuCD andC‐controlsintheArgentinianpopulation(B).Statisticallysignificantdifferencesareshown(two tailedMann‐WhitneyUtest;Krustall‐Wallistest).Horizontalbarsaremedianvalues.Correlation betweentheexpressionofIFNγ andVα24‐Jα18mRNA,inarbitraryunits(AU)intheSpanish population(C)(C‐controls:Spearmanr=0.3115,pvalue=n.s;I‐controls:Spearmanr=0.4265, pvalue=n.s;GFD‐CD:Spearmanr=0.5393,pvalue<0.05;uCD:Spearmanr=0.4323,pvalue<0.05) andintheArgentinianpopulation(D)(C‐controls:Spearmanr=0.5895,pvalue<0.05;uCD: Spearmanr=0.6917,pvalue<0.001). Figure 2. Increased duodenal Vα24-Jα18 mRNA expression in untreated and treated celiac disease patients. Expression levels of Vα24-Jα18 mRNA in duodenal biopsies indicated by the ratio Vα24-Jα18/βactin in arbitrary units (AU), in untreated Celiac Disease (uCD), Gluten Free Diet-CD patients (GFD-CD), inflamed non-CD controls (I-controls) and non-inflamed non-CD controls (C-controls) in the Spanish population (A) and in uCD and C-controls in the Argentinian population (B). Statistically significant differences are shown (two tailed Mann-Whitney U test; Krustall-Wallis test). Horizontal bars are median values. Correlation between the degree of histological lesion (Marsh score) and the expression level of Vα24-Jα18 mRNA in CD patients. (1: Marsh 0-I, 2: Marsh II, 3: Marsh IIIa, 4: Marsh IIIb, 5: Marsh IIIc) There was an increased of Vα24-Jα18 expression and the duodenal increased level of atrophy (Spearman r= 0.063, p < 0.05). GFD-CD patients show a Marsh 0-I, uCD show duodenal atrophy Marsh II to IIIc (C). Nutrients 2015,7,page–page 7  Figure2.IncreasedduodenalVα24‐Jα18mRNAexpressioninuntreatedandtreatedceliacdisease patients.ExpressionlevelsofVα24‐Jα18mRNAinduodenalbiopsiesindicatedbytheratio Vα24‐Jα18/βactininarbitraryunits(AU),inuntreatedCeliacDisease(uCD),GlutenFreeDiet‐CD patients(GFD‐CD),inflamednon‐CDcontrols(I‐controls)andnon‐inflamednon‐CDcontrols (C‐controls)intheSpanishpopulation(A)andinuCDandC‐controlsintheArgentinianpopulation (B).Statisticallysignificantdifferencesareshown(twotailedMann‐WhitneyUtest;Krustall‐Wallis test).Horizontalbarsaremedianvalues.Correlationbetweenthedegreeofhistologicallesion(Marsh score)andtheexpressionlevelofV24‐Jα18mRNAinCDpatients.(1:Marsh0‐I,2:MarshII,3:Marsh IIIa,4:MarshIIIb,5:MarshIIIc)TherewasanincreasedofVα24‐Jα18expressionandtheduodenal increasedlevelofatrophy(Spearmanr=0.063,p<0.05).GFD‐CDpatientsshowaMarsh0‐I,uCD showduodenalatrophyMarshIItoIIIc(C). Figure3.CorrelationbetweenduodenalmRNAexpressionofVα24‐Jα18andIFNγ.Expressionlevels ofIFNγmRNAinduodenalbiopsiesindicatedastheratioIFNγ/βactin,inarbitraryunits(AU),in untreatedCeliacDisease(uCD),GlutenFreeDiet‐CDpatients(GFD‐CD),inflamednon‐CDcontrols (I‐controls)andnon‐inflamednon‐CDcontrols(C‐controls)intheSpanishpopulation(A)andinuCD andC‐controlsintheArgentinianpopulation(B).Statisticallysignificantdifferencesareshown(two tailedMann‐WhitneyUtest;Krustall‐Wallistest).Horizontalbarsaremedianvalues.Correlation betweentheexpressionofIFNγ andVα24‐Jα18mRNA,inarbitraryunits(AU)intheSpanish population(C)(C‐controls:Spearmanr=0.3115,pvalue=n.s;I‐controls:Spearmanr=0.4265, pvalue=n.s;GFD‐CD:Spearmanr=0.5393,pvalue<0.05;uCD:Spearmanr=0.4323,pvalue<0.05) andintheArgentinianpopulation(D)(C‐controls:Spearmanr=0.5895,pvalue<0.05;uCD: Spearmanr=0.6917,pvalue<0.001). Figure 3. Correlation between duodenal mRNA expression of Vα24-Jα18 and IFNγ. Expression levels of IFNγmRNA in duodenal biopsies indicated as the ratio IFNγ/βactin, in arbitrary units (AU), in untreated Celiac Disease (uCD), Gluten Free Diet-CD patients (GFD-CD), inflamed non-CD controls (I-controls) and non-inflamed non-CD controls (C-controls) in the Spanish population (A) and in uCD and C-controls in the Argentinian population (B). Statistically significant differences are shown (two tailed Mann-Whitney U test; Krustall-Wallis test). Horizontal bars are median values. Correlation between the expression of IFNγand Vα24-Jα18 mRNA, in arbitrary units (AU) in the Spanish population (C) (C-controls: Spearman r= 0.3115, pvalue = n.s; I-controls: Spearman r= 0.4265, pvalue = n.s; GFD-CD: Spearman r= 0.5393, pvalue < 0.05; uCD: Spearman r= 0.4323, pvalue < 0.05) and in the Argentinian population (D) (C-controls: Spearman r= 0.5895, pvalue < 0.05; uCD: Spearman r= 0.6917, pvalue < 0.001). 8966 Nutrients 2015,7, 8960–8976 Nutrients 2015,7,page–page 8 3.3.DuodenalVα24‐Jα18andFoxP3mRNALevelsRevealaCeliacDiseaseMolecularProfile WenextstudiedFoxP3mRNAexpressionintheduodenumasanindirectwayofquantifying Tregcellsintissue.CDsamples(bothuCDandGFD‐CD)haddecreasedFoxP3expressioncompared withnon‐inflamednon‐CD‐controls(C‐controls)(uCD,p<0.001;GFD‐CD,p<0.001)(Figure4A). SimilarresultswerefoundonasecondindependentanalysisoftheArgentinianpopulation(uCD, p<0.001)(Figure4B).BecauseCDsampleswerecharacterizedbyincreasedduodenalmRNA expressionofVα24‐Jα18(Figure2A,B)anddecreasedexpressionofFoxP3(Figure4A,B),westudied whetherthejointanalysisofthesetwomoleculescouldhelpustoidentifyaCD‐likemolecularprofile.  Figure4.DuodenalmRNAlevelsofVα24‐Jα18andFoxP3revealaceliacdiseasemolecularprofile. ExpressionlevelsofFoxP3mRNAinduodenalbiopsiesindicatedastheratioFoxP3/βactin,in arbitraryunits(AU)inuntreatedCeliacDisease(uCD),GlutenFreeDiet‐CDpatients(GFD‐CD), inflamednon‐CDcontrols(I‐controls)andnon‐inflamednon‐CDcontrols(C‐controls)intheSpanish population(A)andinuCDandC‐controlsintheArgentinianpopulation(B).Statisticaldifferences areshown(twotailedMann‐WhitneyUtest;Krustall‐Wallistest).Horizontalbarsaremedianvalues. Duodenalmolecularprofile:CorrelationbetweenexpressionlevelsofFoxP3andVα24‐Jα18mRNA,in arbitraryunits(AU)(C‐controls:Spearmanr=0.3495,pvalue<0.05;I‐controls:Spearmanr=−0.4577, pvalue=n.s;GFD‐CD:Spearmanr=−0.4858,pvalue<0.05;uCD:Spearmanr=−0.056,pvalue=n.s)(C). NodifferenceswerefoundbetweensamplesfromcontrolsoruCDpatientsinanyofthe moleculesstudiedinbothpopulations(SpanishandArgentinian).Similarresultswerealsoobtained whenVα24‐Jα18andFoxP3duodenalmRNAexpressionwereindependentlyanalysed.Therefore, bothpopulationsweremergedtoincreasethesamplesizeofuCDandC‐controlgroups.Thejoint analysisofbothduodenalFoxP3andVα24‐Jα18mRNAlevelsallowedustodiscriminatebetween CD(treatedanduntreated),andnon‐CDsamples(inflamedornon‐inflamed)withasensibilityand sensitivityof92%,revealingaCD‐likemolecularprofile(C‐controls:Spearman’sr=0.3495,p<0.05; I‐controls:Spearman’sr=−0.4577,pvalue=n.s;GFD‐CD:Spearman’sr=−0.4858,p<0.05;uCD: Spearman’sr=−0.056,pvalue=n.s)(Figure4C). Figure 4. Duodenal mRNA levels of Vα24-Jα18 and FoxP3 reveal a celiac disease molecular profile. Expression levels of FoxP3 mRNA in duodenal biopsies indicated as the ratio FoxP3/βactin, in arbitrary units (AU) in untreated Celiac Disease (uCD), Gluten Free Diet-CD patients (GFD-CD), inflamed non-CD controls (I-controls) and non-inflamed non-CD controls (C-controls) in the Spanish population (A) and in uCD and C-controls in the Argentinian population (B). Statistical differences are shown (two tailed Mann-Whitney U test; Krustall-Wallis test). Horizontal bars are median values. Duodenal molecular profile: Correlation between expression levels of FoxP3 and Vα24-Jα18 mRNA, in arbitrary units (AU) (C-controls: Spearman r= 0.3495, pvalue < 0.05; I-controls: Spearman r=´0.4577, pvalue = n.s; GFD-CD: Spearman r=´0.4858, pvalue < 0.05; uCD: Spearman r=´0.056, pvalue = n.s) (C). All together, our findings suggest an increase of iNKT cells in the CD duodenum as determined by molecular approaches. Therefore, we next studied the number of iNKT cells within the intraepithelial and lamina propria compartments by flow cytometry. To that end, the IEL profile was first analyzed and compared with previous reports before determining the number of iNKT cells in the CD duodenum. 3.4. Intraepithelial Lymphocytes in the Duodenum from Celiac Disease Patients Total IELs (CD103`CD45`), non-T cells (CD103`CD45`CD3´), TCRγδ`cells (CD103`CD45`CD3`TCRγδ`) and TCRαβ`cells (CD103`CD45`CD3`TCRγδ´) were studied within the intraepithelial compartment (as characterized in Figure 1A) given their relevance as biomarkers in CD diagnosis [9,32]. Untreated CD patients had increased numbers of total IELs (median/IQR; 16.80%/4.80) (Figure 5A,E) together with decreased numbers of non-T cells (3.10%/4.00) (Figure 5B,E). The latter was also true in GFD-CD patients although the total number of IELs did not increase (GFD-CD, 10.40%/4.11) (Figure 5A,E). Within the CD3`subpopulation, both uCD and GFD-CD patients 8967 Nutrients 2015,7, 8960–8976 showed higher numbers of TCRγδ`cells (uCD, 35.78%/11.13; GFD-CD, 32.50%/8.90) (Figure 5C,E), previously described as a distinctive feature of CD patients [9,32]. In support of this, inflamed non-CD controls did not have increased numbers of TCRγδ`cells (I-controls, 7.51%/2.60) (Figure 5C,E) despite the decreased percentage of non-T cells compared with non-inflamed non-CD controls (I-controls, 16.30%/8.12; C-controls, 28.90%/15.75) (Figure 5B,E). However, these patients had increased numbers of TCRαβ`cells (I-controls, 76.20%/8.20) compared with the remainder patient groups (uCD, 59.30%/15.6; GFD-CD, 58.80%/10.90; C-controls, 62.10%/14.5) (Figure 5D,E). Within the intraepithelial compartment, a decreased number of non-T cells together with increased TCRγδ`cells represent a distinctive pattern of CD patients irrespectively of the disease status [32,33]. Reference values of the IEL subpopulations for each patient’s group were expressed as median percentages with the interquartile range (IQR) in Table 3. Table 3. Specificity of intraepithelial lymphocytes profiling in the diagnosis of celiac disease. uCD Median (IQR) GFD-CD Median (IQR) I-controls Median (IQR) C-controls Median (IQR) Total IELs 16.80% (4.80) 10.40% (4.11) 10.70% (3.70) 8.50% (2.60) Non-T cells 3.10% (4.00) 9.34% (3.34) 16.30% (8.12) 28.90% (15.75) TCRγδ`cells 35.78% (11.13) 32.50% (8.90) 7.51% (2.60) 6.44% (2.38) TCRαβ`cells 59.30% (15,6) 58.80% (10,9) 76.20% (8,9) 62.10% (14,5) Percentages of duodenal Intraepithelial lymphocyte (IEL) populations in untreated Celiac Disease (uCD), Gluten Free Diet-CD patients (GFD-CD), inflamed non-CD controls (I-controls) and non-inflamed non-CD controls (C-controls) expressed as median percentages with the interquartile range (IQR). Nutrients 2015,7,page–page 9 Alltogether,ourfindingssuggestanincreaseofiNKTcellsintheCDduodenumasdetermined bymolecularapproaches.Therefore,wenextstudiedthenumberofiNKTcellswithinthe intraepithelialandlaminapropriacompartmentsbyflowcytometry.Tothatend,theIELprofilewas firstanalyzedandcomparedwithpreviousreportsbeforedeterminingthenumberofiNKTcellsin theCDduodenum. 3.4.IntraepithelialLymphocytesintheDuodenumfromCeliacDiseasePatients TotalIELs(CD103 + CD45 + ),non‐Tcells(CD103 + CD45 + CD3 − ),TCRγδ + cells(CD103 + CD45 + CD3 + TCRγδ + )andTCRαβ + cells(CD103 + CD45 + CD3 + TCRγδ − )werestudiedwithintheintraepithelial compartment(ascharacterizedinFigure1A)giventheirrelevanceasbiomarkersinCD diagnosis[9,32]. UntreatedCDpatientshadincreasednumbersoftotalIELs(median/IQR;16.80%/4.80)(Figure5A,E) togetherwithdecreasednumbersofnon‐Tcells(3.10%/4.00)(Figure5B,E).Thelatterwasalsotruein GFD‐CDpatientsalthoughthetotalnumberofIELsdidnotincrease(GFD‐CD,10.40%/4.11) (Figure5A,E).WithintheCD3 + subpopulation,bothuCDandGFD‐CDpatientsshowedhigher numbersofTCRγδ + cells(uCD,35.78%/11.13;GFD‐CD,32.50%/8.90)(Figure5C,E),previously describedasadistinctivefeatureofCDpatients[9,32].Insupportofthis,inflamednon‐CDcontrols didnothaveincreasednumbersofTCRγδ + cells(I‐controls,7.51%/2.60)(Figure5C,E)despitethe decreasedpercentageofnon‐Tcellscomparedwithnon‐inflamednon‐CDcontrols(I‐controls, 16.30%/8.12;C‐controls,28.90%/15.75)(Figure5B,E).However,thesepatientshadincreasednumbers ofTCRαβ + cells(I‐controls,76.20%/8.20)comparedwiththeremainderpatientgroups(uCD, 59.30%/15.6;GFD‐CD,58.80%/10.90;C‐controls,62.10%/14.5)(Figure5D,E). Withintheintraepithelialcompartment,adecreasednumberofnon‐Tcellstogetherwith increasedTCRγδ + cellsrepresentadistinctivepatternofCDpatientsirrespectivelyofthedisease status[32,33].ReferencevaluesoftheIELsubpopulationsforeachpatient’sgroupwereexpressedas medianpercentageswiththeinterquartilerange(IQR)inTable3.  Figure5.Cont. Figure 5. Cont. 8968 Nutrients 2015,7, 8960–8976 40. Dunne, M.R.; Elliott, L.; Hussey, S.; Mahmud, N.; Kelly, J.; Doherty, D.G.; Feighery, C.F. Persistent changes in circulating and intestinal gammadelta T cell subsets, invariant natural killer T cells and mucosal-associated invariant T cells in children and adults with coeliac disease. PLoS ONE 2013,8, e76008. [CrossRef] [PubMed] 41. Leon, F.; Eiras, P.; Roy, G.; Camarero, C. Intestinal intraepithelial lymphocytes and anti-transglutaminase in a screening algorithm for coeliac disease. Gut 2002,50, 740–741. [CrossRef] [PubMed] 42. Fernandez-Banares, F.; Carrasco, A.; Garcia-Puig, R.; Rosinach, M.; Gonzalez, C.; Alsina, M.; Loras, C.; Salas, A.; Viver, J.M.; Esteve, M. Intestinal intraepithelial lymphocyte cytometric pattern is more accurate than subepithelial deposits of anti-tissue transglutaminase iga for the diagnosis of celiac disease in lymphocytic enteritis. PLoS ONE 2014,9, e101249. [CrossRef] [PubMed] 43. Walker, M.M.; Murray, J.A. An update in the diagnosis of coeliac disease. Histopathology 2011,59, 166–179. [CrossRef] [PubMed] 44. Grose, R.H.; Thompson, F.M.; Cummins, A.G. Deficiency of 6b11+ invariant NKT-cells in celiac disease. Dig. Dis. Sci. 2008,53, 1846–1851. [CrossRef] [PubMed] 45. Wingender, G.; Kronenberg, M. Role of NKT cells in the digestive system. Iv. The role of canonical natural killer T cells in mucosal immunity and inflammation. Am. J. Physiol. Gastrointest. Liver Physiol. 2008,294, G1–G8. [CrossRef] [PubMed] 46. Eiras, P.; Roldan, E.; Camarero, C.; Olivares, F.; Bootello, A.; Roy, G. Flow cytometry description of a novel CD3-/CD7+ intraepithelial lymphocyte subset in human duodenal biopsies: Potential diagnostic value in coeliac disease. Cytometry 1998,34, 95–102. [CrossRef] 47. Eiras, P.; Leon, F.; Camarero, C.; Lombardia, M.; Roldan, E.; Bootello, A.; Roy, G. Intestinal intraepithelial lymphocytes contain a CD3CD7+ subset expressing natural killer markers and a singular pattern of adhesion molecules. Scand. J. Immunol. 2000,52, 1–6. [CrossRef] [PubMed] 48. Yu, K.O.; Porcelli, S.A. The diverse functions of CD1D-restricted NKT cells and their potential for immunotherapy. Immunol. Lett. 2005,100, 42–55. [CrossRef] [PubMed] 49. Shimizu, K.; Shinga, J.; Yamasaki, S.; Kawamura, M.; Dorrie, J.; Schaft, N.; Sato, Y.; Iyoda, T.; Fujii, S. Transfer of mRNA encoding invariant NKT cell receptors imparts glycolipid specific responses to T cells and gammadeltat cells. PLoS ONE 2015,10, e0131477. [CrossRef] [PubMed] 50. O’Keeffe, J.; Podbielska, M.; Hogan, E.L. Invariant natural killer T cells and their ligands: Focus on multiple sclerosis. Immunology 2015,145, 468–475. [CrossRef] [PubMed] 51. Schrumpf, E.; Tan, C.; Karlsen, T.H.; Sponheim, J.; Bjorkstrom, N.K.; Sundnes, O.; Alfsnes, K.; Kaser, A.; Jefferson, D.M.; Ueno, Y.; et al. The biliary epithelium presents antigens to and activates natural killer T cells. Hepatology 2015,62, 1249–1259. [CrossRef] [PubMed] 52. Sarra, M.; Cupi, M.L.; Monteleone, I.; Franze, E.; Ronchetti, G.; Di Sabatino, A.; Gentileschi, P.; Franceschilli, L.; Sileri, P.; Sica, G.; et al. IL-15 positively regulates IL-21 production in celiac disease mucosa. Mucosal Immunol. 2013,6, 244–255. [CrossRef] [PubMed] 53. Van Bergen, J.; Mulder, C.J.; Mearin, M.L.; Koning, F. Local communication among mucosal immune cells in patients with celiac disease. Gastroenterology 2015,148, 1187–1194. [CrossRef] [PubMed] 54. Gill, N.; Rosenthal, K.L.; Ashkar, A.A. Nk and nkt cell-independent contribution of interleukin-15 to innate protection against mucosal viral infection. J. Virol. 2005,79, 4470–4478. [CrossRef] [PubMed] 55. Pillai, V.; Ortega, S.B.; Wang, C.K.; Karandikar, N.J. Transient regulatory T-cells: A state attained by all activated human T-cells. Clin. Immunol. 2007,123, 18–29. [CrossRef] [PubMed] 56. Bernardo, D.; Al-Hassi, H.O.; Mann, E.R.; Tee, C.T.; Murugananthan, A.U.; Peake, S.T.; Hart, A.L.; Knight, S.C. T-cell proliferation and forkhead box p3 expression in human T cells are dependent on T-cell density: Physics of a confined space? Hum. Immunol. 2012,73, 223–231. [CrossRef] [PubMed] 57. Gibbons, D.L.; Spencer, J. Mouse and human intestinal immunity: Same ballpark, different players; different rules, same score. Mucosal Immunol. 2011,4, 148–157. [CrossRef] [PubMed] 58. Ebert, L.M.; Tan, B.S.; Browning, J.; Svobodova, S.; Russell, S.E.; Kirkpatrick, N.; Gedye, C.; Moss, D.; Ng, S.P.; MacGregor, D.; et al. The regulatory T cell-associated transcription factor FoxP3 is expressed by tumor cells. Cancer Res. 2008,68, 3001–3009. [CrossRef] [PubMed] 8975 Nutrients 2015,7, 8960–8976 59. Morgan, M.E.; van Bilsen, J.H.; Bakker, A.M.; Heemskerk, B.; Schilham, M.W.; Hartgers, F.C.; Elferink, B.G.; van der Zanden, L.; de Vries, R.R.; Huizinga, T.W.; et al. Expression of FoxP3 mRNA is not confined to CD4`CD25`T regulatory cells in humans. Hum. Immunol. 2005,66, 13–20. [CrossRef] [PubMed] 60. Chen, G.Y.; Chen, C.; Wang, L.; Chang, X.; Zheng, P.; Liu, Y. Cutting edge: Broad expression of the FoxP3 locus in epithelial cells: A caution against early interpretation of fatal inflammatory diseases following in vivo depletion of FoxP3-expressing cells. J. Immunol. 2008,180, 5163–5166. [CrossRef] [PubMed] © 2015 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/). 8976