The Journal of Allergy and Clinical Immunology
Volume 120, Issue 6 , Pages 1292-1300, December 2007

IL-13 involvement in eosinophilic esophagitis: Transcriptome analysis and reversibility with glucocorticoids

  • Carine Blanchard, PhD

      Affiliations

    • Division of Allergy and Immunology, Department of Pediatrics, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio
    • ,
  • Melissa K. Mingler, MS

      Affiliations

    • Division of Allergy and Immunology, Department of Pediatrics, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio
    • ,
  • Maria Vicario, PhD

      Affiliations

    • Division of Allergy and Immunology, Department of Pediatrics, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio
    • Digestive Diseases Research Unit, Department of Gastroenterology, Hospital Universitari Vall d'Hebron, Universtat Autònoma de Barcelona, Barcelona, Spain
    • ,
  • J. Pablo Abonia, MD

      Affiliations

    • Division of Allergy and Immunology, Department of Pediatrics, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio
    • ,
  • Yi Ying Wu, MS

      Affiliations

    • Division of Allergy and Immunology, Department of Pediatrics, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio
    • ,
  • Thomas X. Lu, BS

      Affiliations

    • Division of Allergy and Immunology, Department of Pediatrics, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio
    • ,
  • Margaret H. Collins, MD

      Affiliations

    • Division of Pathology and Laboratory Medicine, Cincinnati Children's Hospital and Medical Center, University of Cincinnati College of Medicine, Cincinnati, Ohio
    • ,
  • Philip E. Putnam, MD

      Affiliations

    • Division of Gastroenterology, Hepatology and Nutrition, Cincinnati Children's Hospital and Medical Center, University of Cincinnati College of Medicine, Cincinnati, Ohio
    • ,
  • Susanne I. Wells, PhD

      Affiliations

    • Division of Hematology/Oncology, Department of Pediatrics, Cincinnati Children's Hospital and Medical Center, University of Cincinnati College of Medicine, Cincinnati, Ohio
    • ,
  • Marc E. Rothenberg, MD, PhD

      Affiliations

    • Division of Allergy and Immunology, Department of Pediatrics, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio
    • Corresponding Author InformationReprint requests: Marc Rothenberg, MD, PhD, Cincinnati Children's Hospital Medical Center, 3333 Burnet Ave, ML 7028, Cincinnati, Ohio 45229.

Received 3 October 2007; received in revised form 12 October 2007; accepted 15 October 2007.

Article Outline

Background

Eosinophilic esophagitis (EE) is an emerging worldwide disease that mimics gastroesophageal reflux disease. Early studies have established that esophageal eosinophilia occurs in association with TH2 allergic responses, and we recently identified an EE-specific esophageal transcriptome that included eotaxin-3.

Objective

We sought to determine the mechanism by which this TH2 response leads to EE.

Methods

Real-time PCR and microarray analysis were performed on RNA extracted from esophageal biopsy specimens and primary esophageal epithelial cell cultures stimulated with IL-13 (0-100 ng/mL). Transient transfections in esophageal cell lines were performed with plasmids containing the luciferase gene driven by eotaxin-3 promoter fragments and modified forms of signal transducer and activator of transcription 6.

Results

The IL-13 mRNA level was markedly increased (16-fold) in esophageal biopsy specimens from patients with EE compared with those from healthy individuals. Furthermore, IL-13 treatment of primary esophageal epithelial cells was sufficient to induce a global-expression transcript profile that remarkably overlapped with the EE-specific esophageal transcriptome. In addition, esophageal epithelial cells markedly produce eotaxin-3 after IL-13 stimulation through a transcriptional mechanism dependent on signal transducer and activator of transcription 6. Lastly, increased IL-13 mRNA levels and the EE transcriptome were largely reversible with glucocorticoid treatment in vivo.

Conclusions

Taken together, we propose that the pathogenesis of EE is mediated by an IL-13–stimulated keratinocyte-derived transcriptome that is largely reversible with corticosteroid treatment. Furthermore, we identify an in vivo IL-13–induced transcriptome that has potential utility for target assessment after anti-IL-13 therapeutics.

Clinical implications

IL-13–induced pathways and genes are fundamental processes in the cause and manifestations of EE; as such, therapeutic agents that interfere with IL-13 might be particularly useful for disease treatment.

Key words: IL-13, eosinophilic esophagitis, eotaxin-3, cytokines, esophageal epithelial cells, keratinocytes

Abbreviations used: EE, Eosinophilic esophagitis, ER, Estrogen receptor, FP, Fluticasone propionate, STAT6, Signal transducer and activator of transcription 6

 

Eosinophilic esophagitis (EE) is an emerging worldwide disease that mimics gastroesophageal reflux disease and can lead to esophageal narrowing and stricture.1 EE is differentiated from gastroesophageal reflux disease by the lack of response to acid suppression, the magnitude of mucosal eosinophilia and epithelial thickening, its male predominance, and a high rate of association with atopy.2, 3, 4

The TH2 cell–derived cytokines IL-4 and IL-13 are produced at increased levels in allergic tissue and are thought to be central regulators of many of the hallmark features of eosinophilic disease.5 Although some preliminary investigations have suggested that circulating T cells and eosinophils overproduce IL-13 in patients with EE,6, 7 the production of IL-13 by esophageal cells, including infiltrating cells, has not been examined. The mechanisms by which IL-13 mediates its effects vary between tissues5, 8, 9 because IL-13 induces dramatically different transcriptional profiles in different cell types.10, 11, 12

All 3 human eotaxin promoters contain signal transducer and activator of transcription 6 (STAT6) consensus-binding motifs, but their relative importance and mode of action differ.13, 14, 15, 16, 17, 18 Interestingly, patients with EE have an esophageal transcriptome characterized by differential expression of 574 genes that is remarkably conserved between individuals, despite their atopic status or sex.19 Notably, eotaxin-3 is the most overexpressed gene within the EE transcriptome (53-fold increase compared with healthy individuals), even though eotaxin-1 and eotaxin-2 are not significantly induced in patients with EE.

In this study we aimed to uncover the molecular mechanisms involved in the development of EE by focusing on the signaling pathway responsible for the induction of the EE-specific transcriptome. Our data frame the theory that the pathogenesis of EE is mediated by an IL-13–stimulated keratinocyte-derived transcriptome that is largely reversible with corticosteroid treatment. In addition, we have defined an in vivo IL-13–associated gene signature with potential value for efficacy assessment of anti-IL-13 therapeutics that are under active clinical testing.

Back to Article Outline

Methods 

Cell culture 

A human esophageal adenocarcinoma cell line and squamous epithelial cells (TE1, TE6, TE7, and TE13) were provided by Dr Hainault (IARC, Lyon, France). These cell lines, originally selected from esophageal tumors and well characterized by Nishihira et al,20, 21 were maintained in RPMI medium (Invitrogen, Carlsbad, Calif) supplemented with 10% FCS and 1% penicillin/streptomycin/amphotericin (Invitrogen). For primary epithelial cell culture, the culture conditions and cytokine treatments are provided in the Methods section in the Online Repository (available at www.jacionline.org). This study was approved by the Institutional Review Board of the Cincinnati Children's Hospital Medical Center.

DNA microarray analysis 

RNA extraction, microarray analysis, and ontological assessment were performed as previously described.19 For identification of the resistant genes, the EE transcriptome gene list was applied to fluticasone propionate (FP)–treated patients, and genes significantly expressed differently (P ≤ .05) between FP-treated and healthy patients were subjected to fold-change filter (≥2). Response to therapy was defined as 1 or fewer eosinophil per high-powered field and no epithelial cell hyperplasia after at least 3 months' treatment with FP; these sets of patients are extensions of the analysis done previously.22

Reporter constructs, expression vectors, and transient transfections 

Plasmids were constructed or obtained and cells were transfected, as described previously.13 Approximately 104 cells per well (24-well plates) were plated on day 1 and transfected with Lipofectamine reagent (Invitrogen), as previously described.13

Flow cytometric analysis 

Flow cytometric analysis of interleukin receptor chains (human IL-4Rα, human IL-13Rα1, and human IL-13Rα2) was performed as previously described.13 Briefly, the TE-7 cell line was incubated with 1 μg of mAbs, anti-human IL-4Rα, anti-human IL-13Rα1 (mAb IgG1; Diaclone, Stamford, Conn), or anti-human IL-13Rα2 or IgG1 isotype and 0.4 μg of fluorescein isothiocyanate–labeled secondary antibody. Cells were then analyzed for fluorescence by using single-color flow cytometry with a flow cytometer FACScalibur and analyzed with FlowJo software (TreeStar, Inc, Ashland, Ore).

ELISA protocol for IL-13, eotaxin-1, eoataxin-2, and eotaxin-3 quantification 

Eotaxin-1, eotaxin-2, and eotaxin-3 Duo set ELISAs were performed according to the manufacturer's instruction (R&D Systems, Minneapolis, Minn), as previously described.19, 23 The IL-13 Quantikine Kit (R&D Systems) was used to quantify IL-13 protein levels. The detection limits were 70, 200, 200 and 62 pg/mL for eotaxin-1, eotaxin-2, eotaxin-3, and IL-13, respectively.

Statistical analysis 

Data are expressed as means ± SD. Statistical significance comparing 2 different treatments or groups was determined by using the Student t test (normal distribution equal variance), the Welch t test (normal distribution unequal variances), the Mann-Whitney test (nonparametric test, 2 groups), or the Kruskal-Wallis test, followed by a Dunn multiple comparison test (nonparametric test, ≥3 groups) with Prism 4 Software.

Back to Article Outline

Results 

IL-13 expression in patients with EE 

To establish the participation of IL-13 (and/or IL-4) in EE pathogenesis, we wanted first to establish whether IL-13 was overproduced in the esophageal tissue of patients with EE. By using real-time PCR analysis, there was a 16-fold increase in IL-13 mRNA in patients with EE compared with that seen in healthy patients (defined as individuals with no gastrointestinal pathology) (Fig 1, A). Interestingly, the IL-4 mRNA level was not significantly increased in our patients with EE compared with that seen in healthy patients, although a subgroup of patients showed an increase in IL-4 levels (Fig 1, B). Notably, there was a correlation between IL-13 mRNA and eotaxin-3 expression (r2 = 0.49; P < .05).

  • View full-size image.
  • Fig 1. 

    IL-13 and IL-4 mRNA expression in biopsy samples from healthy (NL) subjects and patients with EE. The expression of IL-13 (A) and IL-4 (B) is shown. Each mRNA value is normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA expression from the same sample and is expressed as a fold increase. The black lines represent the mean value in each group. P values were calculated by using the Mann-Whitney U test (2 groups; n = 8-9 and 13-21 subjects for the healthy and EE groups, respectively).

IL-13–induced esophageal epithelial genes 

Primary esophageal epithelial cells from the esophagi of patients with EE were cultured and stimulated with 100 ng/mL IL-13 for 48 hours to identify IL-13–induced esophageal epithelial genes. The mRNA was subjected to global transcript-expression profile analysis and normalized pairwise to unstimulated controls. Genes were filtered to fold change in biopsy specimens from patients with EE compared with those from healthy patients and IL-13–stimulated epithelial cells. A highly significant positive correlation (P < .0001, Spearman) between modified genes in biopsy specimens from patients with EE and in IL-13–stimulated keratinocytes was observed (Fig 2, A). Eotaxin-3, cadherin-26, and TNF-α–induced protein 6 were upregulated in both specimens from patients with EE and IL-13–stimulated keratinocytes. Mast cell genes (carboxypeptidase A3), eosinophil genes (Charcot Leyden Crystal protein) and lymphocyte genes (immunoglobulin chains) were upregulated more than 10-fold in patients with EE compared with values in healthy patients19 but were not upregulated in IL-13–stimulated epithelial cells (Fig 2, A). In IL-13–stimulated cells, 4698 and 952 genes were significantly modified by using different stringencies of statistical comparisons (P < .05 and P < .01, respectively) compared with unstimulated cells. There were 1333 genes modified by more than 1.5-fold (780 upregulated and 553 downregulated) and 371 genes modified by 2-fold or greater (255 genes were upregulated and 116 were downregulated; Fig 2, B and C, and Table E1 in the Online Repository at www.jacionline.org). Ontological analysis of the 780 upregulated genes revealed that the most significant biologic processes involved pathways that regulated cell cycle (P = 7.04 × 10−6), response to external stimulus (P = 1.33 × 10−5), response to wounding (P = 1.77 × 10−5), and cell proliferation (P = 2.90 × 10−5). In contrast, downregulated genes were involved in ectoderm development (P = 3.01 × 10−7), epidermis development (P = 7.28 × 10−7), tissue development (P = 2.53 × 10−5), and keratinization (P = 2.63 × 10−5). Genes that were upregulated 5-fold or greater and downregulated 4-fold or greater in primary esophageal keratinocytes stimulated with IL-13 (100 ng/mL, 48 hours) are shown in Fig 2, B.

  • View full-size image.
  • Fig 2. 

    Gene expression analysis in primary esophageal cells after IL-13 stimulation and comparison with the EE transcript signature. A, The 54,765 genes of the HG-U133 chip were subjected to fold-change filter in patients with EE versus healthy subjects and IL-13–stimulated primary cell cultures versus unstimulated cells. Spearman correlation and linear regression were calculated. B, The list displays 33 transcripts that were upregulated 5-fold or greater and 5 transcripts that were downregulated 4-fold or greater compared with unstimulated cells. C, The genes modified by 1.5-fold or greater on average in IL-13–stimulated cells (100 ng/mL) are presented in a heat diagram in 3 primary-culture patient biopsy specimens (1, 2, and 3), unstimulated and stimulated. Upregulated genes are shown in red, and downregulated genes are shown in blue. The magnitude of the gene changes is proportional to the darkness of the color. D, The fold increase of eotaxin-3 mRNA expression compared with the untreated value was quantified by means of real-time PCR. E, Eotaxin-3 released in the culture supernatant is expressed in nanograms per milliliter. Results are presented as means ± range and are representative of experiments performed in 5 different patients.

We were interested in determining whether the IL-13–regulated gene signature overlapped with the EE transcriptome.19 We therefore compared the IL-13–induced gene transcript profile with the EE-specific transcriptome (Fig 2, A-C). Notably, 126 of the IL-13–induced genes (22% of the EE transcriptome, P < .05) overlapped with the EE transcriptome. Interestingly, the number one gene overexpressed in IL-13–stimulated keratinocytes was eotaxin-3, which was upregulated 279-fold (Fig 2, B). Remarkably, eotaxin-1 and eotaxin-2 were not overexpressed in IL-13–treated primary keratinocyte cultures.

IL-13–induced eotaxin-3 expression ex vivo 

Primary keratinocytes were stimulated with 0 to 100 ng/mL human IL-13 for 48 hours to further test whether IL-13 induces eotaxin-3 mRNA expression in the esophagi of patients with EE. Eotaxin expression was studied by means of real-time PCR (Fig 2, D). Eotaxin-3 expression was induced in a dose-dependent manner after IL-13 stimulation. Remarkably, there was a 1000- and 10,000-fold increase of eotaxin-3 mRNA expression 48 hours after 10 and 100 ng/mL IL-13, respectively. Although not detectable at baseline, IL-13 (100 ng/mL) induced a dramatic release of eotaxin-3 protein (Fig 2, E) into the supernatant (6.03 ± 0.8 ng/mL), representing at least a 30-fold increase over the detection limit (200 pg/mL). Importantly, eotaxin-1 and eotaxin-2 mRNA and protein were not overexpressed under these conditions (data not shown).

Eotaxin-3 expression in esophageal epithelial cell lines after IL-13 stimulation 

To investigate the molecular mechanisms involved in IL-13–induced eotaxin-3 expression in keratinocytes, we examined human esophageal epithelial cell lines (TE-1, TE-6, TE-7, and TE-13). First, by using RT-PCR, we demonstrated that the receptor chains of IL-13 (IL-13Rα1, IL-13Rα2, and IL4Rα) are expressed by these cell lines (Fig 3, A). The respective protein products were detected by means of FACS analysis in TE-7 esophageal epithelial cells (Fig 3, B). All cell lines were subsequently stimulated with increasing concentrations of human IL-13 for 0 to 48 hours (Fig 3, C). Eotaxin-3 mRNA expression was increased in a dose-dependent manner, although to varying degrees between the cell lines. Eotaxin-3 mRNA expression was increased by 4-, 8-, 77-, and 1007-fold in TE-1, TE-6, TE-7, and TE-13 cells, respectively, after IL-13 treatment at 100 ng/mL. Baseline eotaxin-3 protein expression was less than the detection limit of 200 pg/mL but was overexpressed in the cell contents and in the supernatant of IL-13–stimulated TE-7 cells in a dose- and time-dependent fashion (Fig 3, D and E). After IL-13 stimulation (100 ng/mL for 24 hours), 2.5 ± 0.7 ng/mL and 0.55 ± 0.37 ng/mL eotaxin-3 was released in the supernatants of TE-7 and TE-13 cells, respectively. Notably, eotaxin-1 and eotaxin-2 expression levels were less than or at the detection limit of the real-time PCR and were not overexpressed in these cell lines after IL-13 stimulation (data not shown).

  • View full-size image.
  • Fig 3. 

    IL-13 receptor chain expression in esophageal cells and eotaxin-3 production by esophageal cell lines after IL-13 stimulation. A, The TE-1, TE-6, TE-7, and TE-13 esophageal cell lines were subjected to PCR analysis for IL-4Rα, IL-13Rα1, IL-13Rα2, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA expression. B, Flow cytometric analysis of IL-4R, IL-13Rα1, and IL-13Rα2 chain expression in the TE-7 cell line (dark lines). Controls were performed with an irrelevant IgG1 (gray lines). C, The TE-1, TE-6, TE-7, and TE-13 esophageal cell lines were stimulated for 24 hours with IL-13 (0, 1, 10, and 100 ng/mL). The fold increase of eotaxin-3 mRNA expression compared with that of the untreated cells is shown. D and E, TE-7 cells were cultured with IL-13 (0, 1, 10, and 100 ng/mL) for 1, 6, 12, 24, and 48 hours. Eotaxin-3 protein released in the culture supernatant was quantified by means of ELISA. Results are presented as means ± range and are representative of at least 3 experiments performed in triplicate.

Eotaxin-3 promoter activity 

Activated STAT6 has been shown to bind to consensus sites located at −698 and −89 bp in skin and kidney fibroblasts and epithelial cell lines of the lung and colon.13, 14, 15, 24 To determine the STAT6 element or elements involved in the eotaxin-3 stimulation observed in esophageal cell lines, we transiently transfected the TE-7 cell line with the full-length eotaxin luciferase reporter plasmid and truncated versions. There was a significant increase of 2.7 ± 0.8–fold and 10.8 ± 2.2–fold of luciferase activity with the full-length reporter plasmid at IL-13 doses of 10 and 100 ng/mL, respectively (Fig 4, A). Similar results were observed in the other TE cell lines (data not shown). The TE-7 cell line was transfected with truncated promoter constructs or with a full-length promoter containing specific mutations within the STAT6-binding sites to map the relevant cis-acting promoter sequences (Fig 4, B). After IL-13 stimulation, the construct P100 (containing a truncated proximal STAT6 site) had no activity, whereas the promoter containing 800 or 500 bp had full IL-13–induced activity. Furthermore, the construct containing the mutated −89 binding site had no increase in luciferase activity, demonstrating that the proximal STAT6-responsive element at base pair −89 was required for IL-13–induced eotaxin-3 promoter activity.

  • View full-size image.
  • Fig 4. 

    Human eotaxin-3 promoter activity after IL-13 stimulation and the role of STAT6. A, TE-7 cells were transfected with pGL3 basic containing the eotaxin-3 promoter (P800) and phRLTK. Cells were stimulated with IL-13 (0, 1, 10, and 100 ng/mL) B, TE-7 cells were transfected with pGL3 basic containing different lengths of the eotaxin-3 promoter (P800, P500, and P100) and promoters containing mutations in the STAT6-responsive elements (MUT1, MUT2, and MUT1&2). C, The TE-7 cells were cotransfected with P800 and a dominant-negative form of STAT6 (DNSTAT6) or the empty vector (EV). D, TE-7 cells were cotransfected with P800 and the expression vector containing STAT6:ER. The cells were stimulated with 4-hydroxytamoxifene (4HT; 10 μmol/L). Results are presented as the ratio of the luciferase firefly/Renilla activities. E, Esophageal keratinocytes (TE-7) were pretreated with IL-13 (0 or 100 ng/mL) and actinomycin D (Actino; 0 or 10 μmol/L) for 0 to 48 hours. Results are presented as a percentage of eotaxin-3 mRNA compared with time 0 hours (100%; black and gray dashed lines for media and IL-13, respectively). Trend lines (black and gray lines for media and IL-13, respectively) were calculated.

STAT6-dependent mechanism 

Cells were first cotransfected with the eotaxin-3 reporter and a dominant-negative STAT6-expressing vector or empty control vector to definitively implicate STAT6 in eotaxin-3 induction (Fig 4, C). The overexpression of the dominant-negative STAT6 dramatically decreased (P < .05) activity of the IL-13–induced eotaxin-3 promoter. The cells were also cotransfected with the eotaxin-3 promoter construct and an estrogen-inducible form of STAT6 (fusion between STAT6 and estrogen receptor [ER], STAT6:ER).25 Subsequently, addition of 4-hydroxytamoxifene to the culture medium (which allows dimerization of the modified ER protein fused to STAT626) resulted in a significant (P < .01) induction of the eotaxin-3 promoter luciferase activity within 24 hours (Fig 4, D).

Eotaxin-3 mRNA stability 

Chemokine mRNA expression can be highly dependent on posttranscriptional mechanisms, such as RNA stability. Experiments with the RNA polymerase II inhibitor actinomycin-D revealed that the stability of eotaxin-3 mRNA was not significantly different between IL-13–treated and nontreated TE-7 cells (Fig 4, E), and the half-life of eotaxin-3 in esophageal keratinocytes was determined to be 5.7 ± 2.3 hours and not significantly modified by IL-13 treatment (Fig 4, E). Additionally, the eotaxin-3 3′ untranslated region sequence was subcloned downstream of an SV40 promoter–driven luciferase gene in the pGL3 reporter plasmid. Transfections were performed in the presence or absence of IL-13. No statistically significant increase of luciferase activity after IL-13 stimulation was observed (data not shown).

Identification of glucocorticoid-reversible genes in patients with EE 

We have previously reported that FP induces EE disease remission compared with placebo treatment.22 We aimed to identify whether IL-13 overexpression could be normalized in patients with EE with successful anti-inflammatory intervention. Indeed, IL-13 mRNA levels were significantly (P < .01) reduced in EE responders compared with untreated specimens, and expression levels after treatment were comparable with the levels detected in control biopsy specimens (Fig 5, A). Similarly, eotaxin-3 mRNA was normalized in FP-responder patients (P < .01; Fig 5, B). We next aimed to determine whether the EE transcriptome was also reversible in patients with EE successfully treated with glucocorticoids. Remarkably, 98% of the EE transcriptome was reversed to expression levels detected in biopsy specimens of healthy patients (Fig 5, C). The reversible genes include cell-specific transcripts, including eosinophil, mast cell, lymphocyte, fibroblast, and epithelial genes. In addition, genes involved in cellular recruitment and cell proliferation were also reversible. Interestingly, although the biopsy specimens had no abnormal microscopic features, 12 genes were still dysregulated (Fig 5, D). Among these 12 dysregulated genes, uroplakin and cadherin-26 remained upregulated, and desmoglein remained downregulated. Indeed, real-time PCR demonstrated a residual upregulation of cadherin-26 in FP-treated patients with EE (P < .05; Fig 5, E).

  • View full-size image.
  • Fig 5. 

    Effect of glucocorticoids on the EE transcriptome and resistant genes. A and B, The expression of IL-13 (Fig 5, A) and eotaxin-3 (Fig 5, B) mRNA is shown in healthy subjects (NL), patients with EE, and patients with EE treated successfully with FP (n = 8-9, 13-19, and 6-8 subjects for the NL, EE, and FP groups, respectively). C, Total mRNA was subjected to microarray analysis. Upregulated genes are shown in red, and downregulated genes are shown in blue. Each column represents a separate patient (NL, EE, and FP Rx), and each line represents a gene. D, Genes that are resistant to glucocorticoid therapy are shown with their Affymetrix accession numbers and their fold change in patients with EE and in treated patients with EE. E, Expression of cadherin-26 (CDH26) was quantified by means of real-time PCR. Each data point corresponds to a separate individual (n = 9, 11, and 7 subjects for the NL, EE, and FP groups, respectively). P values were calculated using Kruskal-Wallis tests (3 groups).

Back to Article Outline

Discussion 

In this study we demonstrated that a large number of EE-associated genes are directly induced by IL-13 in esophageal keratinocytes, and we therefore implicate IL-13 as a major regulator of the keratinocyte pathways involved in EE. In EE, the esophageal tissue undergoes changes marked by an abnormal accumulation of eosinophils, mast cells,7, 13, 27 and lymphocytes7, 13; epithelial cell hyperplasia; elongation of the papillae (endothelial cells and fibroblasts); and intensive lamina propria remodeling (likely involving fibroblasts).3, 28, 29, 30, 31, 32 In our minimalist model system, the stimulation of keratinocytes with IL-13 is able to partially reproduce the EE transcriptome, indicating that this cell type is likely to largely account for the abnormal response seen in endoscopic biopsy specimens. Based on this striking overlap, we support a model in which IL-13–induced gene expression in keratinocytes makes an important contribution to EE. This model, however, does not exclude indirect and paracrine effects.

In a previous study using microarray analysis, IL-4 and IL-13 mRNA were neither detected nor upregulated in patients with EE; however, several other TH2-inducible molecules were still detected.19 Using a highly sensitive method (real-time PCR) relative to microarray analysis, we demonstrate that IL-13 mRNA is indeed overexpressed in patients with EE (16-fold). This is consistent with previous studies showing an increase in IL-13 levels in stimulated PBMCs and eosinophils of patients with EE.7, 33, 34 Mast cells, basophils, eosinophils, lymphocytes, and smooth muscle cells are potent IL-13–producing cells6, 7, 34, 35; however, the exact cellular source or sources of IL-13 in the esophageal biopsy specimen remains to be elucidated. IL-4 was overexpressed in less than 50% of patients with EE, suggesting a more important involvement for IL-13 in the mucosa for most patients.

In the esophagus, enhanced expression of eotaxin-3 transcript by IL-13 is likely operational through the transcription factor STAT6, with no significant affect on mRNA half-life based on our studies (Fig 4, E). Our results are consistent with a recent study in a colonic epithelial cell line13 but contrast with the literature describing IL-13–induced eotaxin-1 mRNA stabilization in airway epithelial cells; this suggests that different mechanisms can be used for the regulation of distinct eotaxin family members in different cell types.36 It is notable that the EE transcriptome and IL-13–induced genes do not include eotaxin-1 and eotaxin-2, despite the presence of STAT6-binding sites in both of these genes. Taken together, these results suggest that the regulation of eotaxin-3 occurs differently from that of the other eotaxins and that keratinocytes use a regulatory pathway unique from that of other cells.

Topical FP therapy has been shown to improve clinical symptoms, as well as endoscopic and microscopic features of EE.22, 37, 38 In this study we uncover the mechanism involved by demonstrating that successful FP treatment reverses the molecular signature of EE. Although topical glucocorticoids are known to have anti-inflammatory effects, they do not universally reduce levels of all cytokines,39 and therefore it was important to determine the effect of topical fluticasone on the EE transcriptome. Notably, glucocorticoid treatment was associated with reduced IL-13 and eotaxin-3 production, indicating that the classic features of this TH2-associated pathway in the esophagus are largely reversible. These reversible genes include cell-specific transcripts from eosinophils, mast cells, lymphocytes, fibroblasts, and epithelial cells, as well as chemoattractants, growth factors, and molecules involved in cell proliferation. These results are consistent with the decrease in eosinophils, mast cells, and epithelial hyperplasia after therapy.22, 37 Interestingly, although the treated biopsy specimens appear microscopically normal, the transcriptome still contains a small number of dysregulated genes. The residual expression of these genes in successfully treated patients with EE suggests that these genes (cadherin-26, desmoglein 1, and uroplakin 1B) might be part of the primary constitutive genetic defect inherent to the epithelium (modifying permeability or elasticity) or have a reduced propensity to respond to glucocorticoid treatment. Of note, eotaxin-3 mRNA and protein levels were only partially decreased by glucocorticoids in IL-13–stimulated esophageal keratinocytes (data not shown). Of interest, the residual expression markers have potential clinical value because they might indeed serve as diagnostic criteria, irrespective of the degree of tissue inflammation (and the expression of the rest of the EE transcriptome). Additionally, these resistant genes might also help in the understanding of the chronic and relapsing nature of the disease.40

In summary, our results provide new insight into the molecular pathogenesis of EE. We propose that disease pathogenesis involves a glucocorticoid-reversible IL-13–induced keratinocyte transcriptome that includes eotaxin-3. These results underscore the potential value of new therapeutics interfering with the IL-13/eotaxin-3/CCR3 axis41; it is predicted that such agents would not only limit eosinophil accumulation but also the keratinocyte proliferation characteristic of EE. Therapeutic agents that interfere with IL-13 signaling are under active development for asthma. It has not escaped our attention that our newly defined set of IL-13–induced EE transcripts has potential value in the testing of clinical reagents that block IL-13 in patients. Not only do we propose that anti-IL-13 therapeutics be examined in patients with EE because of their potential clinical benefit but also because the EE transcriptome provides a robust means to molecularly monitor drug efficacy and mechanism of action.

Back to Article Outline

 

We thank Andrea Lippelman for assistance with the preparation of the manuscript, Elisabeth Jones for technical assistance, and Drs Fred Finkelman and Mitchell Cohen for review of the manuscript.

Back to Article Outline

Appendix. Supplementary data 

Download file

download text

Online Repository.

Back to Article Outline

References 

  1. Furuta GT, Liacouras CA, Collins HC, Gupta SK, Justinich C, Putnam PE, et al. Eosinophilic esophagitis in children and adults: a systematic review and consensus recommendations for diagnosis and treatment. Gastroenterology. 2007;133:1342–1363
  2. Noel RJ, Putnam PE, Rothenberg ME. Eosinophilic esophagitis. N Engl J Med. 2004;351:940–941
  3. Blanchard C, Wang N, Rothenberg ME. Eosinophilic esophagitis: pathogenesis, genetics, and therapy. J Allergy Clin Immunol. 2006;118:1054–1059
  4. Rothenberg ME. Eosinophilic gastrointestinal disorders (EGID). J Allergy Clin Immunol. 2004;113:11–29
  5. Hershey GK. IL-13 receptors and signaling pathways: an evolving web. J Allergy Clin Immunol. 2003;111:677–691
  6. Schmid-Grendelmeier P, Altznauer F, Fischer B, Bizer C, Straumann A, Menz G, et al. Eosinophils express functional IL-13 in eosinophilic inflammatory diseases. J Immunol. 2002;169:1021–1027
  7. Straumann A, Bauer M, Fischer B, Blaser K, Simon HU. Idiopathic eosinophilic esophagitis is associated with a T(H)2-type allergic inflammatory response. J Allergy Clin Immunol. 2001;108:954–961
  8. Shim YM, Zhu Z, Zheng T, Lee CG, Homer RJ, Ma B, et al. Role of 5-lipoxygenase in IL-13-induced pulmonary inflammation and remodeling. J Immunol. 2006;177:1918–1924
  9. Ingram JL, Antao-Menezes A, Mangum JB, Lyght O, Lee PJ, Elias JA, et al. Opposing actions of Stat1 and Stat6 on IL-13-induced up-regulation of early growth response-1 and platelet-derived growth factor ligands in pulmonary fibroblasts. J Immunol. 2006;177:4141–4148
  10. Lee JH, Kaminski N, Dolganov G, Grunig G, Koth L, Solomon C, et al. Interleukin-13 induces dramatically different transcriptional programs in three human airway cell types. Am J Respir Cell Mol Biol. 2001;25:474–485
  11. Homer RJ, Zhu Z, Cohn L, Lee CG, White WI, Chen S, et al. Differential expression of chitinases identify subsets of murine airway epithelial cells in allergic inflammation. Am J Physiol Lung Cell Mol Physiol. 2006;291:L502–L511
  12. Zheng T, Zhu Z, Liu W, Lee CG, Chen Q, Homer RJ, et al. Cytokine regulation of IL-13Ralpha2 and IL-13Ralpha1 in vivo and in vitro. J Allergy Clin Immunol. 2003;111:720–728
  13. Blanchard C, Durual S, Estienne M, Emami S, Vasseur S, Cuber JC. Eotaxin-3/CCL26 gene expression in intestinal epithelial cells is up-regulated by interleukin-4 and interleukin-13 via the signal transducer and activator of transcription 6. Int J Biochem Cell Biol. 2005;37:2559–2573
  14. Hebenstreit D, Luft P, Schmiedlechner A, Duschl A, Horejs-Hoeck J. SOCS-1 and SOCS-3 inhibit IL-4 and IL-13 induced activation of Eotaxin-3/CCL26 gene expression in HEK293 cells. Mol Immunol. 2005;42:295–303
  15. Hoeck J, Woisetschlager M. Activation of eotaxin-3/CCLl26 gene expression in human dermal fibroblasts is mediated by STAT6. J Immunol. 2001;167:3216–3222
  16. Hoeck J, Woisetschlager M. STAT6 mediates eotaxin-1 expression in IL-4 or TNF-alpha-induced fibroblasts. J Immunol. 2001;166:4507–4515
  17. Matsukura S, Stellato C, Georas SN, Casolaro V, Plitt JR, Miura K, et al. Interleukin-13 upregulates eotaxin expression in airway epithelial cells by a STAT6-dependent mechanism. Am J Respir Cell Mol Biol. 2001;24:755–761
  18. Chen W, Khurana Hershey GK. Signal transducer and activator of transcription signals in allergic disease. J Allergy Clin Immunol. 2007;119:529–543
  19. Blanchard C, Wang N, Stringer KF, Mishra A, Fulkerson PC, Abonia JP, et al. Eotaxin-3 and a uniquely conserved gene-expression profile in eosinophilic esophagitis. J Clin Invest. 2006;116:536–547
  20. Nishihira T, Hashimoto Y, Katayama M, Mori S, Kuroki T. Molecular and cellular features of esophageal cancer cells. J Cancer Res Clin Oncol. 1993;119:441–449
  21. Nishihira T, Kasai M, Mori S, Watanabe T, Kuriya Y, Suda M, et al. Characteristics of two cell lines (TE-1 and TE-2) derived from human squamous cell carcinoma of the esophagus. Gann. 1979;70:575–584
  22. Konikoff MR, Noel RJ, Blanchard C, Kirby C, Jameson SC, Buckmeier BK, et al. A Randomized, double-blind, placebo-controlled trial of fluticasone propionate for pediatric eosinophilic esophagitis. Gastroenterology. 2006;131:1381–1391
  23. Konikoff MR, Blanchard C, Kirby C, Buckmeier BK, Cohen MB, Heubi JE, et al. Potential of blood eosinophils, eosinophil-derived neurotoxin, and eotaxin-3 as biomarkers of eosinophilic esophagitis. Clin Gastroenterol Hepatol. 2006;4:1328–1336
  24. Yuan Q, Campanella GS, Colvin RA, Hamilos DL, Jones KJ, Mathew A, et al. Membrane-bound eotaxin-3 mediates eosinophil transepithelial migration in IL-4-stimulated epithelial cells. Eur J Immunol. 2006;36:2700–2714
  25. Kamogawa Y, Lee HJ, Johnston JA, McMahon M, O'Garra A, Arai N. A conditionally active form of STAT6 can mimic certain effects of IL-4. J Immunol. 1998;161:1074–1077
  26. Pritchard CA, Samuels ML, Bosch E, McMahon M. Conditionally oncogenic forms of the A-Raf and B-Raf protein kinases display different biological and biochemical properties in NIH 3T3 cells. Mol Cell Biol. 1995;15:6430–6442
  27. Kirsch R, Bokhary R, Marcon MA, Cutz E. Activated mucosal mast cells differentiate eosinophilic (allergic) esophagitis from gastroesophageal reflux disease. J Pediatr Gastroenterol Nutr. 2007;44:20–26
  28. Sant'Anna AM, Rolland S, Fournet JC, Yazbeck S, Drouin E. Eosinophilic esophagitis in children: symptoms, histology and pH probe results. J Pediatr Gastroenterol Nutr. 2004;39:373–377
  29. Furuta GT. Eosinophilic esophagitis: an emerging clinicopathologic entity. Curr Allergy Asthma Rep. 2002;2:67–72
  30. Dauer EH, Freese DK, El-Youssef M, Thompson DM. Clinical characteristics of eosinophilic esophagitis in children. Ann Otol Rhinol Laryngol. 2005;114:827–833
  31. Parfitt JR, Gregor JC, Suskin NG, Jawa HA, Driman DK. Eosinophilic esophagitis in adults: distinguishing features from gastroesophageal reflux disease: a study of 41 patients. Mod Pathol. 2006;19:90–96
  32. Liacouras CA, Ruchelli E. Eosinophilic esophagitis. Curr Opin Pediatr. 2004;16:560–566
  33. Straumann A, Kristl J, Conus S, Vassina E, Spichtin HP, Beglinger C, et al. Cytokine expression in healthy and inflamed mucosa: probing the role of eosinophils in the digestive tract. Inflamm Bowel Dis. 2005;11:720–726
  34. Yamazaki K, Murray JA, Arora AS, Alexander JA, Smyrk TC, Butterfield JH, et al. Allergen-specific in vitro cytokine production in adult patients with eosinophilic esophagitis. Dig Dis Sci. 2006;51:1934–1941
  35. Grunstein MM, Hakonarson H, Leiter J, Chen M, Whelan R, Grunstein JS, et al. IL-13-dependent autocrine signaling mediates altered responsiveness of IgE-sensitized airway smooth muscle. Am J Physiol Lung Cell Mol Physiol. 2002;282:L520–L528
  36. Atasoy U, Curry SL, Lopez de Silanes I, Shyu AB, Casolaro V, Gorospe M, et al. Regulation of eotaxin gene expression by TNF-alpha and IL-4 through mRNA stabilization: involvement of the RNA-binding protein HuR. J Immunol. 2003;171:4369–4378
  37. Noel RJ, Putnam PE, Collins MH, Assa'ad AH, Guajardo JR, Jameson SC, et al. Clinical and immunopathologic effects of swallowed fluticasone for eosinophilic esophagitis. Clin Gastroenterol Hepatol. 2004;2:568–575
  38. Teitelbaum JE, Fox VL, Twarog FJ, Nurko S, Antonioli D, Gleich G, et al. Eosinophilic esophagitis in children: immunopathological analysis and response to fluticasone propionate. Gastroenterology. 2002;122:1216–1225
  39. Adcock IM, Ito K, Barnes PJ. Glucocorticoids: effects on gene transcription. Proc Am Thorac Soc. 2004;1:247–254
  40. Assa'ad AH, Putnam PE, Collins MH, Akers RM, Jameson SC, Kirby CL, et al. Pediatric patients with eosinophilic esophagitis: an 8-year follow-up. J Allergy Clin Immunol. 2007;119:731–738
  41. Klion AD, Bochner BS, Gleich GJ, Nutman TB, Rothenberg ME, Simon HU, et al. Approaches to the treatment of hypereosinophilic syndromes: a workshop summary report. J Allergy Clin Immunol. 2006;117:1292–1302

 Supported in part by National Institutes of Health grants AI070235 and AI45898 (M.E.R.), the Food Allergy and Anaphylaxis Network (M.E.R.), the Campaign Urging Research for Eosinophil Disorders, the Buckeye Foundation (M.E.R.), the Food Allergy Project (M.E.R.), the American Heart Association 0625296B (C.B.), the Thrasher Research Funds NR-0014 (C.B.), the Instituto Carlos III Fondo de Investigación Sanitaria CD05/00060 (M.V.), Public Health Service grant CA102357 (S.I.W.), the Translational Research Initiative, Cincinnati Children's Hospital Medical Center, and the DHC (NIDDK 064403).

 Disclosure of potential conflict of interest: M. E. Rothenberg has consulting arrangements with Ception Therapeutics and Merck, owns stock in Ception Therapeutics, and is on the speakers' bureau for Merck. The rest of the authors have declared that they have no conflict of interest.

 Transcript profiling is available online at http://cypher.cchmc.org:1104. The reader should login as a guest and select “HG-U133 genome”; experiments and gene lists are located in the folder named MRothenberg/blanchard et al.

PII: S0091-6749(07)01974-4

doi:10.1016/j.jaci.2007.10.024

The Journal of Allergy and Clinical Immunology
Volume 120, Issue 6 , Pages 1292-1300, December 2007

Article Tools

* Image Usage & Resolution