We have previously proposed that the pathogenesis of eosinophilic esophagitis (EE) is mediated by an IL-13–driven epithelial cell response associated with marked gene dysregulation including eotaxin-3 overproduction. In this study, we compared epithelial responses between healthy patients and those with EE, aiming to uncover molecular explanations for EE pathogenesis. Esophageal epithelial cells could be maintained for up to five passages, with 67% and 62% of cell lines reaching confluence in healthy controls and EE cases, respectively. Both sets of epithelial cells avidly responded to IL-13 at similar levels as assessed by eotaxin-3 production. Acidic pH increased cellular release of eotaxin-3 (4.6 ± 1.98 ng/ml versus 12.46 ± 2.90 ng/ml at pH 7.4 and 4, respectively; p < 0.05). Numerous epidermal differentiation complex (EDC) genes, such as filaggrin and SPRR3, were downregulated both in IL-13–stimulated esophageal epithelial cells and in EE biopsies specimens compared with healthy controls. Whereas the filaggrin loss of function mutation 2282del4 was overrepresented in EE compared with control individuals (6.1% versus 1.3% respectively; p = 0.0172), the decreased filaggrin expression was uniformly seen in all EE cases in vivo. Indeed, expression of the EDC genes filaggrin and involucrin was strongly decreased directly by IL-13. These results establish that the epithelial response in EE involves a cooperative interaction between IL-13 and expression of EDC genes.

Eosinophilic esophagitis (EE) is a complex atopic disorder of the esophageal mucosa characterized by allergen-induced eosinophilic infiltration and epithelial cell hyperplasia (1). Yet the growth characteristics of normal (NL) and EE epithelial cells in vitro have not been studied, and there are limited data regarding the intrinsic properties of esophageal epithelial cells in patients with EE (1). Patients with EE typically exhibit symptoms that mimic gastroesophageal reflux disease (GERD), but GERD and EE are distinguished by the lack of histologic response to acid suppression therapy in EE (2). The distinction between GERD and EE is not always clear, as GERD can have substantial esophageal eosinophilia and patients with EE sometimes report clinical responses to acid neutralization therapy.

There is a striking overlap between the mucosal structure of the skin and the esophagus (3, 4). During the terminal differentiation of epidermal keratinocytes, cells undergo desquamation resulting in cornified cells at the surface. These dead cells compose a major part of the epithelial barrier and are continuously renewed. Interestingly, the human esophageal epithelium is not normally cornified and expresses differentiation markers distinct from the skin (58). For example, involucrin is expressed in early-differentiating cells in the esophagus and in the late-differentiating cells in the skin (8). Proliferation/differentiation gene expression have been largely studied in the context of the skin (9) in normal and pathologic conditions (10), whereas markers of differentiation of the esophageal epithelium have not been extensively studied. During skin keratinocyte differentiation, keratin filament organization is modified by the keratin-binding protein filaggrin in stratified epithelia (10, 11). Association of filaggrin or related proteins of the epidermal differentiation complex (EDC) genes clustered on chromosome 1q21 (12), (e.g., involucrin, newly identified component of the EDC, loricrin, small proline-rich region protein [SPRR]) with keratin prevents the proteolytic destruction of keratin during epidermal cell terminal differentiation (1315), which in turn forms an important part of the epithelial barrier of cornified cells (10, 13, 15, 16). The importance of this pathway in the development of atopy is underscored by the marked effect of loss of function mutations in the filaggrin gene, which predispose to atopic dermatitis (AD), a disease that coexists in ∼50% of EE patients. Indeed both EE and AD share molecular characteristics including food sensitization. Notably, defects in epithelial barrier function associated with filaggrin mutations have been associated with AD (1723); such mutations are only present in 0–5% of control individuals and 9–27% of AD individuals, depending upon the study (17, 24).

We have recently proposed that IL-13 is a key cytokine in EE disease pathogenesis. In particular, esophageal epithelial cells express all components of the IL-13 receptor including IL-4Rα, IL-13Rα1, and IL-13Rα2 (25). In addition, IL-13 induces prominent dysregulation of gene expression in the esophageal epithelium including marked overexpression of eotaxin-3, the most strongly induced IL-13 target gene, which is also highly overexpressed in vivo in the EE esophageal transcriptome (25). Furthermore, intratracheal IL-13 induces features of experimental EE in mice, and IL-13– and STAT6- deficient mice are protected from the development of experimental EE (2628). A recent study has shown that in patients with AD, some EDC genes might be downregulated by IL-13 through a STAT6-dependent mechanism (29). In our initial analysis of the EE transcriptome, we have recently noted that filaggrin mRNA was dramatically decreased in the esophagus of EE compared with NL biopsy specimens using microarray analysis (30), suggesting abnormal epithelial differentiation in EE. In this study, we aimed to determine whether there is an intrinsic defect in epithelial cell responses in EE and to further determine which features of the disease can be mediated by IL-13. We report a profound dysregulation of EDC gene expression, identify the effects of IL-13 on EDC gene expression and the presence of gene variants in the EDC gene filaggrin in patients with EE.

Primary esophageal epithelial cells were grown as previously described (25). Human esophageal epithelial cell line (TE) and squamous epithelial cells were provided by Dr. Hainault (International Agency for Research on Cancer, Lyon, France) and were maintained in RPMI 1640 medium (Invitrogen, Carlsbad, CA) supplemented with 10% FCS (Atlanta, J0138, Advantage) and 1% penicillin/streptomycin/amphotericin (Invitrogen). In preliminary studies, we optimized the culture conditions by modifying the source of the FBS, which had a potential effect on cell growth and responses (Supplemental Fig. 1). For primary cell culture, one or two distal esophageal biopsies from patients with EE were collected during routine endoscopy and with informed consent as approved by the institutional review board. The eosinophil level in each patient was quantified and corresponds to the maximum or peak eosinophils found in one high-power field (hpf). All sections of every biopsy specimen were assessed and counted. The samples were cultured in modified F-media (3:1 F-12/DMEM) supplemented with 5%, FBS (Atlanta, J0138, Advantage), adenine (24.2 μg/ml), cholera toxin (10−4 μM), insulin (5 μg/ml), hydrocortisone (0.4 μg/ml), and human epidermal growth factor (10 ng/ml) in the presence of penicillin, streptomycin, and amphotericin (Invitrogen). Tissue was first digested twice with trypsin, followed by trypsin neutralization in F-media without epidermal growth factor. The cells were then plated onto 1–5 × 105 NIH 3T3 J2 feeder cells irradiated with 6000 rad. The next day, the media was exchanged with F-media containing epidermal growth factor (10 ng/ml) and fresh media were added every other day. NL and EE patient-derived cells were cultured for 2 wk in this way, and feeder fibroblasts were then removed by differential trypsinization. Cells were stimulated with IL-13 (0, 1, 10, 100 ng/ml) obtained from Peprotech (Rocky Hill, NJ), as described (25). For acidification experiments, a pH 4 solution of HCl/NaCl (300 mosmol/kg H2O) or a pH 7.4 solution of NaCl (300 mosmol/kg H2O) was added for 4 h. After IL-13 stimulation, 250 μl saline or HCl/saline was added for 4 h. Supernatants were harvested and neutralized, and volumes were adjusted before eotaxin-3 quantification. Viable cells were counted using a hematocytometer and trypan blue staining.

Total cellular RNA was extracted with Trizol (Invitrogen) according to the manufacturer’s instructions. The samples were DNase treated, and first strand cDNA was synthesized using Superscript II (Invitrogen). Total RNA from biopsy samples was extracted using the Mini RNA Extraction Kit (Qiagen, Valencia, CA), and reverse transcription was performed using Iscript (Bio-Rad, Hercules, CA). Real-time PCR was performed by rapid-cycling using the IQ5 (Bio-Rad) SYBR mix (Bio-Rad) as a ready-to-use reaction mix according to the manufacturer’s instructions. The following primers set were: GAPDH (350 bp), 5′-TGGAAATCCCATCACCATCT-3′ and 5′-GTCTTCTGGGTGGCAGTGAT-3′; Filaggrin (125 bp), 5′-GGGAAGTTATCTTTTCCTGTC-3′ and 5′-GATGTGCTAGCCCTGATGTTG-3′; Involucrin (378 bp), 5′-GTTCCTCCTCCAGTCAATACCCATC-3′ and 5′-CTTCATTCCCAGTTGCTCATCTCTC-3′; Eotaxin-3 (125 bp), 5′-GTAACTCTGGGAGGAAACACCCTCTCC-3′ and 5′-AACTCCGAAACAATTGTGACTCAGCTG-3′. PCR products were sequenced using the Cincinnati Children’s Hospital Medical Center (Cincinnati, OH) sequencing core facility.

The probe sets of the EE transcriptome and IL-13–induced genes were subjected to David ontology 2.0 DAVID (database for annotation, visualization, and integrated discovery) and expression analysis systematic explorer analyses, a Web-based (http://david.abcc.ncifcrf.gov/) application (Laboratory of Immunopathogenesis and Bioinformatics, SAIC-Frederick, Frederick, MD) that allows access to a relational database of functional annotations (31, 32). Chromosome locations of the transcripts were obtained, and p values were calculated based on the total number of genes present on each chromosome. Microarray data are available online at www.ncbi.nlm.nih.gov/projects/geo/, the accession number is GSE8853 (25).

Patient DNA was isolated from saliva, peripheral blood leukocytes, or oral buccal swab samples. The history of allergic diseases (asthma, AD, atopic rhinitis) was recorded in our EE population. EE was defined as ≥24 eosinophils per hpf in at least 1 hpf of the esophageal biopsy as previously described (33). For R501X, the following primers were used 5′-CTG GAG GAA GAC AAG GAT CG-3′ and 5′-TTG TCT GCT TGC ACT TCT GG-3′ to amplify a 245-bp fragment as previously described (34). The PCR product was digested for 1 h with NlaIII at 37°C, and electrophoresis was performed on a 4% agarose gel with ethidium bromide. An allele with no mutation digested into two fragments of 213 bp and 32 bp. An allele with the R501X mutation digested into three fragments of 176 bp, 37 bp, and 32 bp. For 2282del4, the following primers 5′-AAT AGG TCT GGA CAC TCA GGT-3′ and 5′-GGG AGG ACT CAG ACT GTT T-3′ were used to amplify an 811-bp fragment as previously described (35). Two fragments (671 bp and 136 bp) were obtained for the mutated alleles. The nonmutated allele did not contain the restriction site. The PCR products were digested for 1 h with DraIII at 37°C and electrophoresis was performed on a 1% agarose gel with ethidium bromide as previously described (1723).

Statistical significance comparing different treatments or groups was determined by the Student t test (normal distribution equal variance), Welch t test (normal distribution, unequal variances), Mann-Whitney U test (non parametric test, two groups), or ANOVA and Kruskal-Wallis test followed by a Dunn’s multiple comparison test (nonparametric test, three groups or more) using Prism 5 Software (GraphPad, San Diego, CA). For the case control study, χ2, odds ratio (OR), and 95% confidence interval were calculated using Prism 5 Software (GraphPad).

We aimed to determine whether primary esophageal epithelial cells from NL and EE patients had similar growth characteristics. The percentage of cells able to reach 80–100% confluence by 35 d was similar with 67 and 62% of NL and EE cell lines, respectively (Fig. 1A). No differences were found in the number of cells at confluence or after 35 d in culture with 1.8 ± 1.3 × 106 and 1.3 ± 1.0 × 106 in NL and EE cells, respectively (Fig. 1B). An average of 19 ± 8 d was sufficient to reach confluence (80–100%) in the NL group compared with 21 ± 8 d in the EE group (Fig. 1C).

FIGURE 1.

Growth characteristics and response to IL-13 of primary esophageal epithelial cells from healthy patients and those with EE. Cells were grown in modified F-media for a maximum of 35 d. A, The percentage of cells that reach 80–100% confluence is shown in the NL and EE groups. B, The number of cells at confluence or after 35 d of culture. C, The number of days to reach 80–100% confluence. D, The fold increase in eotaxin-3 mRNA expression following IL-13 stimulation at different doses is shown; note the Log10 scale. Data points are average ± SD. E, EE- and NL-derived cells were stimulated for 48 h with IL-13 (100 ng/ml); then the media was acidified or heparinized for 4 h and eotaxin-3 protein was quantified in the supernatant. Bars indicate mean values ± SD; *p < 0.05. ns, nonsignificant.

FIGURE 1.

Growth characteristics and response to IL-13 of primary esophageal epithelial cells from healthy patients and those with EE. Cells were grown in modified F-media for a maximum of 35 d. A, The percentage of cells that reach 80–100% confluence is shown in the NL and EE groups. B, The number of cells at confluence or after 35 d of culture. C, The number of days to reach 80–100% confluence. D, The fold increase in eotaxin-3 mRNA expression following IL-13 stimulation at different doses is shown; note the Log10 scale. Data points are average ± SD. E, EE- and NL-derived cells were stimulated for 48 h with IL-13 (100 ng/ml); then the media was acidified or heparinized for 4 h and eotaxin-3 protein was quantified in the supernatant. Bars indicate mean values ± SD; *p < 0.05. ns, nonsignificant.

Close modal

To test whether epithelial cells from NL and EE individuals have the same propensity to respond to IL-13, cells were treated with IL-13 (0–100 ng/ml). The fold change of eotaxin-3 mRNA expression was 281 ± 513 and 64 ± 84-fold increase at 1 ng/ml, 7783 ± 14,689 and 2893 ± 2381-fold increase with IL-13 at 10 ng/ml and 34,174 ± 23,974 and 22,910 ± 24,389-fold increase with IL-13 at 100 ng/ml in NL and EE cells, respectively (Fig. 1D). Eotaxin-3 secretion was dependent on the culture conditions, especially the type of serum present (Supplemental Fig. 1). In addition, eotaxin-3 secretion into the supernatant was affected by proton concentrations. In particular, 4.6 ± 2.0 ng/ml and 12.4 ± 2.9 ng/ml were released into the supernatant of EE-derived cells stimulated with IL-13 (100 ng/ml) when cells were incubated 4 h at pH 7.4 and 4, respectively, although cell viability was not affected (data not shown). In addition, eotaxin-3 release was increased by heparin, as 12.9 ± 3.8 ng/ml eotaxin-3 was obtained after heparin treatment compared with the 4.6 ± 2.0 ng/ml in the absence of heparin. Similar results were obtained in primary epithelial cell cultures derived from NL individuals (Fig. 1E). These results suggest that the esophageal responses in EE may be a cell-extrinsic phenomenon in the presence of IL-13 and microenvironmental components, such as the levels of protons and heparin, affect eotaxin-3 levels.

The EE transcriptome, defined as the 574 probes significantly different (p < 0.001) in esophageal biopsies of EE patients compared with NL patients (30), was analyzed for the chromosomal locations of the respective genes. Chromosome 1 was the most significantly (p = 0.0002) affected chromosome with 48 genes altered (corresponding to 61 probes, or 10.6% of the EE transcriptome; Fig. 2A). We noted the modulation of several EDC genes (such as filaggrin) clustered on chromosome 1q21 (12), a region linked with atopy (36). Using microarray results in NL and EE patient biopsies, we studied the expression pattern of the EDC genes at the location 1q21 (Fig. 2B). Most of the genes in this cluster showed significant decreased expression or trends toward decreased expression (Fig. 2B). Filaggrin mRNA was downregulated 16-fold in EE compared with normal individuals (Fig. 3A). We verified our microarray findings using real-time PCR by examining normal and EE patients to determine the relative expression of filaggrin in different disease states. Using a cohort of 144 individuals, filaggrin mRNA level was downregulated by 16-fold in patients with active EE compared with healthy patients. The esophageal specific SPRR gene esophagin (SPRR3) was highly expressed in the esophagus and significantly decreased 6-fold in patients with EE compared with healthy patients (Fig. 3B). In contrast, the EDC marker involucrin was not significantly downregulated in patients with EE compared with healthy patients (0.58 [0.017 - 0.086] and 0.13 [0.057 - 0.49], respectively; median [25–75 interquartile]; Fig. 3C). These data show that the esophagus of patients with EE has dysregulated expression of select EDC genes, including filaggrin and SPRR3.

FIGURE 2.

Chromosome involvement in the EE transcriptome and physical map of the EDC gene cluster on the chromosome 1. A, The EE transcriptome was assigned to chromosomes using DAVID ontology. The number of genes and the p value are presented. B, A schematic of the physical map of the 1q21 gene cluster and the corresponding microarray expression profile in healthy patients and those with EE. *p < 0.05; **p < 0.005.

FIGURE 2.

Chromosome involvement in the EE transcriptome and physical map of the EDC gene cluster on the chromosome 1. A, The EE transcriptome was assigned to chromosomes using DAVID ontology. The number of genes and the p value are presented. B, A schematic of the physical map of the 1q21 gene cluster and the corresponding microarray expression profile in healthy patients and those with EE. *p < 0.05; **p < 0.005.

Close modal
FIGURE 3.

Esophageal expression of filaggrin mRNA. A, Filaggrin expression was assessed by real-time PCR and normalized to GAPDH in controls (NL), active EE cases (EE active; eosinophil level ≥24), EE cases with an intermediate eosinophilia level (EE int; eosinophils level from 1–23) and inactive EE patients who responded successfully to their therapy (inactive; eosinophil level = 0); n = 48, 21, 18, and 9 individuals in EE active, EE intermediate, EE inactive and NL groups, respectively. Each data point corresponds to a patient, and the lines show the median with interquartile range. C and D, Involucrin and SPRR3 expression were assessed by real-time PCR and normalized to GAPDH as described in A. Each data point corresponds to a patient and the lines show the median with interquartile range.

FIGURE 3.

Esophageal expression of filaggrin mRNA. A, Filaggrin expression was assessed by real-time PCR and normalized to GAPDH in controls (NL), active EE cases (EE active; eosinophil level ≥24), EE cases with an intermediate eosinophilia level (EE int; eosinophils level from 1–23) and inactive EE patients who responded successfully to their therapy (inactive; eosinophil level = 0); n = 48, 21, 18, and 9 individuals in EE active, EE intermediate, EE inactive and NL groups, respectively. Each data point corresponds to a patient, and the lines show the median with interquartile range. C and D, Involucrin and SPRR3 expression were assessed by real-time PCR and normalized to GAPDH as described in A. Each data point corresponds to a patient and the lines show the median with interquartile range.

Close modal

We examined filaggrin expression in treated patients with EE to gain molecular insight into the reversibility of the EDC gene expression changes (Fig. 3A). In patients with EE who responded to therapy (glucocorticoids and/or diet modification), the filaggrin expression level was not significantly different compared with the control level and was significantly (p < 0.01) different compared with active EE. The same trend as filaggrin expression was observed with SPRR3 gene expression; no significant decrease in involucrin gene expression in healthy patients versus patients with active or inactive EE was observed (Fig. 3B, 3C). These results show that abnormal EDC gene expression in EE is reversible following successful therapy.

We propose that the observed decrease in filaggrin mRNA levels could reduce filaggrin function and cause associated disease phenotypes; this is consistent with the finding that filaggrin levels normalized after successful therapy. However, two loss-of-function genetic variants (R501X and 2282del4) in the filaggrin gene have recently been associated with AD; these may not affect the level of filaggrin mRNA, but instead the resulting inactive truncated proteins may compromise epidermal integrity and barrier function. Interestingly, no significant difference in filaggrin mRNA expression was observed between allergic and nonallergic patients with active EE (Fig. 4; Supplemental Table II).

FIGURE 4.

Filaggrin expression in patients with active EE with or without allergic symptoms or sensitization. Filaggrin expression was assessed by real-time PCR and normalized to GAPDH in nonallergic (n = 7) and allergic patients with EE (n = 25). Each data point corresponds to a patient and the lines show the median with the interquartile range.

FIGURE 4.

Filaggrin expression in patients with active EE with or without allergic symptoms or sensitization. Filaggrin expression was assessed by real-time PCR and normalized to GAPDH in nonallergic (n = 7) and allergic patients with EE (n = 25). Each data point corresponds to a patient and the lines show the median with the interquartile range.

Close modal

We genotyped 164 controls and 365 patients with EE for these two mutations. Reliable genotype data for 2282del4 were obtained in 329 of the 365 patients with EE and 157 of all the 164 controls. We found 6.1% heterozygotes for 2282del4 in the EE group and 1.3% in the control group, corresponding to 3% and 0.6% allele frequency in the EE and control group, respectively (Table I). Of the patients with heterozygote EE with the deletion, 60% had a history of eczema and 70% of food allergy (Supplemental Table I). Significant associations (p = 0.0172; OR, 5.016; and p = 0.0185; OR, 4.89) were found between this deletion (genotype and allele frequency, respectively) and EE (Table I). Reliable genotype data for R105X were obtained in 339 of the patients with 365 patients genotyped and all the 164 controls (Table I). We identified 4.4% heterozygotes for R501X in the EE group and 3% in the control group corresponding to 2.2% and 1.5% allele frequency in the EE and control group, respectively (Table I). No significant association was found between this single nucleotide polymorphism and EE. No overlap of mutations was found between heterozygote patients; a total of 11% of our EE population had one or the other mutation compared with 4.3% in the control population (p = 0.0363, Table II). The prevalence of atopy (Table III) was similar across the EE R105X positive, the EE 2282del4 positive, and EE WT groups (86.6%, 85.0%, and 83.3%, respectively). The prevalence of AD was 33%, 60%, and 44% (p = 0.11) across the EE R105X positive, the EE 2282del4 positive, and EE WT groups, respectively. Interestingly, 2282del4 mutation was still associated with EE when EE patients without any history of AD were analyzed (p = 0.0315; Table III). As such, while filaggrin mutations (2282del4 and R501X) were increased in EE compared with control individuals; they were present in only 11% of patients with EE, suggesting that germline mutations in the filaggrin gene are not the primary etiology of filaggrin reduction.

Table I.
2282del4 and R501X mutations in the EE and control population
Deletion 2282del4Patients with EE (n = 329)Controls (n = 157)p Value
Genotype frequency    
 WT 309 (93.9%) 155 (98.7%) p = 0.0172  
 Het 20 (6.1%) 2 (1.3%) OR [CI], 5.0 [1.157-21.74] 
Allele frequency    
 No deletion 638 (97.0%) 312 (99.4%) p = 0.0185 / 
 Deletion 20 (3.0%) 2 (0.6%) OR [CI], 4.89 [1.135- 21.06] 
Deletion 2282del4Patients with EE (n = 329)Controls (n = 157)p Value
Genotype frequency    
 WT 309 (93.9%) 155 (98.7%) p = 0.0172  
 Het 20 (6.1%) 2 (1.3%) OR [CI], 5.0 [1.157-21.74] 
Allele frequency    
 No deletion 638 (97.0%) 312 (99.4%) p = 0.0185 / 
 Deletion 20 (3.0%) 2 (0.6%) OR [CI], 4.89 [1.135- 21.06] 
SNP R501XPatients with EE (n = 339)Controls (n = 164)p Value
CC 324 (96.2%) 159 (97.0%) p = 0.4591 
CT 15 (4.4%) 5 (3.0%)  
663 (97.8%) 323 (98.5%) p = 0.4637 
15 (2.2%) 5 (1.5%)  
SNP R501XPatients with EE (n = 339)Controls (n = 164)p Value
CC 324 (96.2%) 159 (97.0%) p = 0.4591 
CT 15 (4.4%) 5 (3.0%)  
663 (97.8%) 323 (98.5%) p = 0.4637 
15 (2.2%) 5 (1.5%)  

SNP, single nucleotide polymorphism.

Table II.
Combined analysis of R501X and 2282del4 mutation in EE and controls
MutationEE Population (n = 365)Controls (n = 164)p Value
R501X 15 (4.7%) 5 (3.0%) NS 
2282del4 20 (6.3%) 2 (1.2%) NS 
R501X and 2282del4 0 (0%) 0 (0%) NS 
R501X or 2282del4 35 (11.0%) 7 (4.3%) p = 0.0363 
   OR [CI], 2.379 [1.033-5.475] 
Unknowna R501X 26 (7.1%) 0 (0%) NS 
Unknown 2282del4 36 (9.9%) 7 (4.3%) NS 
MutationEE Population (n = 365)Controls (n = 164)p Value
R501X 15 (4.7%) 5 (3.0%) NS 
2282del4 20 (6.3%) 2 (1.2%) NS 
R501X and 2282del4 0 (0%) 0 (0%) NS 
R501X or 2282del4 35 (11.0%) 7 (4.3%) p = 0.0363 
   OR [CI], 2.379 [1.033-5.475] 
Unknowna R501X 26 (7.1%) 0 (0%) NS 
Unknown 2282del4 36 (9.9%) 7 (4.3%) NS 
a

Genotype not determined.

Table III.
Frequency of AD and genotypes in the EE patients
R501X Positive2282del4 PositiveWTp Value
EE without AD 10 131a p = 0.0315 (for 2282Del4) 
Controls 157 OR [CI], 4.794 [1.00-23.98] 
R501X Positive2282del4 PositiveWTp Value
EE without AD 10 131a p = 0.0315 (for 2282Del4) 
Controls 157 OR [CI], 4.794 [1.00-23.98] 
R501X Positive 
(n = 15)2282del4 Positive
(n = 20)WT
(n = 283)p Value
ADb 5 (33.3%) 12 (60%) 102 (44.3%)a NS 
Skin prick test positivec 7 (46.7%) 10 (50.0%) 72 (69.9%)d NS 
History of atopic diseasee 13 (86.6%) 17 (85.0%) 210 (83.3%)f NS 
R501X Positive 
(n = 15)2282del4 Positive
(n = 20)WT
(n = 283)p Value
ADb 5 (33.3%) 12 (60%) 102 (44.3%)a NS 
Skin prick test positivec 7 (46.7%) 10 (50.0%) 72 (69.9%)d NS 
History of atopic diseasee 13 (86.6%) 17 (85.0%) 210 (83.3%)f NS 
a

Information unavailable for 50 of the genotyped patients.

b

Past or present history of allergic dermatitis.

c

History of at least one positive skin prick testing to food or aeroallergen.

d

Information unavailable for 180 of the genotyped patients.

e

History of AD, atopic rhinitis, asthma, or positive skin-prick test.

f

Information unavailable for 31 of the genotyped patients.

We next aimed to investigate the effect of IL-13 on epithelial cell growth and differentiation. Chronic exposure of esophageal epithelial cells to IL-13 (100 ng/ml) for 9 d increased cell growth modestly (mean ± SD, 44.5 ± 3.1 to 65.2 ± 11.4 × 104 cells; p < 0.05; Fig. 5A). In contrast, a similar stimulation with IL-13 for 48 h did not increase the cell number (data not shown). We aimed to determine whether the expression of the EDC genes such as involucrin, SPRRs, and filaggrin were downregulated by IL-13 (Fig. 5B–D). Microarray analysis revealed that filaggrin, involucrin, and SPRR4, 1A, and 3 were significantly downregulated in IL-13 treated cells. Using real time PCR, IL-13 (100 ng/ml) decreased filaggrin and involucrin mRNA expression by more than 9-fold and 5-fold in primary esophageal epithelial cells, respectively (Fig. 5C, 5D). Interestingly, pretreatment with glucocorticoids was not able to inhibit IL-13 downregulation of filaggrin mRNA (data not shown). Using a dominant negative form of STAT6, we examined the role of STAT6 in the decreased expression observed after IL-13 stimulation. Expression of a dominant negative form of STAT6 partially rescued the IL-13 mediated repression of involucrin (Fig. 5E), suggesting that STAT6 functionally contributes to IL-13–regulated involucrin expression in esophageal epithelial cells.

FIGURE 5.

Physical map and gene expression of the EDC gene cluster on the chromosome 1 in IL-13–stimulated primary esophageal epithelial cells. A, The effect of chronic exposure of IL-13 on esophageal epithelial cell number is shown. Cells (10% confluence) were stimulated every 3 d with IL-13 (0, 10, 100 ng/ml) for 9 d (80–90% confluence). Cell numbers are presented as mean ± SD. The results are representative of at least three independent experiments. B, Schematics of the physical map of the 1q21 gene cluster and the microarray expression profile primary esophageal epithelial cells stimulated with IL-13. B and C, Esophageal primary epithelial cells from patients with EE were stimulated with IL-13 (100 ng/ml) for 48 h. Total mRNA was extracted and subjected to real-time RT-PCR for involucrin (B) and filaggrin (C). D, The esophageal epithelial cell line TE-7 (<10% confluence) was transfected with a dominant-negative (DN) form of STAT6 or the empty vector (EV). Total mRNA was extracted and submitted to real-time RT-PCR for involucrin mRNA quantification. Results are normalized to a housekeeping gene (GAPDH). Bars are mean ± SD. Experiments were performed in triplicate. *p < 0.05. DN, dominant-negative; EV, empty vector.

FIGURE 5.

Physical map and gene expression of the EDC gene cluster on the chromosome 1 in IL-13–stimulated primary esophageal epithelial cells. A, The effect of chronic exposure of IL-13 on esophageal epithelial cell number is shown. Cells (10% confluence) were stimulated every 3 d with IL-13 (0, 10, 100 ng/ml) for 9 d (80–90% confluence). Cell numbers are presented as mean ± SD. The results are representative of at least three independent experiments. B, Schematics of the physical map of the 1q21 gene cluster and the microarray expression profile primary esophageal epithelial cells stimulated with IL-13. B and C, Esophageal primary epithelial cells from patients with EE were stimulated with IL-13 (100 ng/ml) for 48 h. Total mRNA was extracted and subjected to real-time RT-PCR for involucrin (B) and filaggrin (C). D, The esophageal epithelial cell line TE-7 (<10% confluence) was transfected with a dominant-negative (DN) form of STAT6 or the empty vector (EV). Total mRNA was extracted and submitted to real-time RT-PCR for involucrin mRNA quantification. Results are normalized to a housekeeping gene (GAPDH). Bars are mean ± SD. Experiments were performed in triplicate. *p < 0.05. DN, dominant-negative; EV, empty vector.

Close modal

In this study, we compared epithelial responses between healthy patients and those with EE to uncover molecular explanations for EE pathogenesis. Esophageal epithelial cells from both sets grew with comparable characteristics and avidly responded to IL-13 at similar levels. Numerous EDC genes, such as filaggrin, and SPRR3 were downregulated in both IL-13–stimulated esophageal epithelial cells and EE compared with NL biopsies specimens. Whereas a filaggrin loss of function mutation (2282del4) was overrepresented in a subset of patients with EE compared with control individuals, the acquired decrease in filaggrin expression was uniformly seen in all patients with EE in vivo, suggesting a generalized importance. The EDC genes filaggrin and involucrin expression were differentially increased during esophageal epithelial cell differentiation (Supplemental Fig. 2) and strongly decreased by IL-13.

In our cell culture condition, NL- and EE-derived biopsy specimens had similar growth characteristics. These results suggest that the esophageal hyperplasia observed in EE may result from an extrinsic phenomenon. It has been shown that esophageal epithelial cells differentially express a multitude of genes in EE compared with healthy controls (30). Whereas the upregulated genes include proliferation markers, such as Ki67, the downregulated genes include members of the EDC (33, 37). In patients with EE, there is severe epithelial cell hyperplasia, and the number of epithelial cells with early esophageal differentiation markers (e.g., involucrin) is increased. In addition, there is a decrease in occurrence of well-differentiating cells. This finding may explain partially why the involucrin gene, expressed in early differentiating cells, is not significantly decreased in EE. In contrast, filaggrin, expressed in well differentiating cells, is highly downregulated in patients with EE. Although both are downregulated by IL-13, the increased number of involucrin-expressing cells might be responsible for the involucrin mRNA expression level not significantly changing in EE biopsy samples. We have previously shown a strong overlap between the IL-13–stimulated esophageal epithelial cell transcriptome and the EE gene signature, suggesting a major involvement of IL-13 in EE (25, 30). Chronic exposure of esophageal epithelial cells to IL-13 induces a modest increase in cell number in vitro. IL-13 has been shown in vivo to induce esophageal epithelial hyperplasia in mice (38). Our results also demonstrate that NL and EE epithelial cells both avidly respond to IL-13 by overexpressing eotaxin-3. Although these results are representative of only one gene, they suggest that the IL-13/IL-13 receptor/STAT6 pathway is not dramatically altered in NL versus EE. Our data are thus in line with the increase in IL-13 levels in EE biopsies previously reported (25) and argue against the pathogenic involvement of keratinocyte-intrinsic defects. In addition, filaggrin is restored to normal levels when patients respond to glucocorticoids. This glucocorticoid effect is most likely caused by the decrease of IL-13 in vivo, as in vitro experiments have shown that glucocorticoids do not modify IL-13–induced filaggrin expression in esophageal epithelial cells (data not shown). Murine models have shown the potentially important role of IL-13 in EE, as IL-13–deficient mice have a decrease in eosinophil infiltration. Similar observations are seen in IL-4/IL-13 double-deficient and STAT6 deficient mice (27). EE is thus a multifactorial disease that certainly involves IL-13, (at least in part), but also numerous other components (genetic, immunologic, or environmental) that may partially overlap in their role and function in EE pathogenesis.

Acid reflux and reflux diseases are confounders in EE diagnosis, and acid neutralization resolves esophageal eosinophilia in some patients (2). The release of eotaxin-3 by protons may explain why proton pump inhibitors often have partial clinical effects in EE (1, 2). Protons are known to modify barrier function of the squamous epithelium by disrupting intercellular desmosomes. Interestingly, epithelial permeability to protons differs between healthy subjects and patients with active GERD (39). It is interesting to note that eotaxin-3 is associated with the outer surface of the cell membrane, possibly via CCR3, which has been reported to be expressed by skin keratinocytes (40, 41), Notably, heparin releases bound eotaxin-3 from the cell surface. Chemokines are known to bind to glycosaminoglycan side chains of proteoglycans, such as heparin, and heparin potentiates eotaxin-induced eosinophil recruitment in vivo (42). The role of heparin in eotaxin-3 release has been previously demonstrated in bronchoalveolar epithelial cells (43) and in blood where the level of eotaxin-3 in heparin-plasma is higher compared with EDTA plasma (33, 44). These findings emphasize the potential role of mast cell-derived heparin in eosinophil recruitment in EE. Consistent with the ability of external factors to regulate eotaxin-3 release, eotaxin-3 recovery in the cell media was dependent on the type of sera used. These results demonstrate that the external components in the microenvironment likely regulate eotaxin-3 levels and the propensity of eosinophils to infiltrate the tissue.

EDC genes and filaggrin are strongly downregulated in patients with EE (30). It has also been shown that a strong association exists between filaggrin mutations and AD (45). Because a large part of the EE population (∼43%) has eczema or a history of eczema, we hypothesized that these mutations would be associated with EE. Although the filaggrin mutations would be expected to be enriched in an atopic population, the prevalence of atopic diseases was the same in EE R105X positive, EE 2282del4 positive, and EE WT groups. In addition, EE patients without AD had a significantly increased frequency of the 2282del4 mutation compared with controls (Table III), suggesting that the association of the 2282del4 mutation with EE was not dependent on the high rate of AD in EE. The significant association found between EE and 2282del4 suggests enhanced percutaneous sensitization in EE (21). It is interesting to note that the frequency of the R501X was below that expected in such a highly atopic population, and this needs further investigation. Our results demonstrate that mutations in filaggrin may separately contribute to EE, but are present in only a small subset of the EE population despite the profound downregulation of filaggrin gene expression seen in nearly all patients.

In conclusion, we report that esophageal epithelial cells from patients with EE are functionally normal compared with cells isolated from control individuals, as assessed by growth properties ex vivo. IL-13 responsiveness based on eotaxin-3 production. Although a filaggrin loss of function mutation (2282del4) was overrepresented in EE compared with control individuals, the acquired decrease in filaggrin expression was uniformly seen in all EE cases in vivo. Indeed, EDC gene (filaggrin and involucrin) expressions were strongly decreased directly by IL-13. As such, these results establish that the primary defect in epithelial responses in EE is not intrinsic to the epithelium, rather it is likely secondary to the effects of IL-13, which regulates expression of EDC genes. Our results support a model of coordinated interaction between IL-13 and epithelial differentiation cluster genes in EE.

We thank Lawanda Bryant, Katherine Henderson, Kiran Mudambi, and Tova Holowinko for assistance.

Disclosures M.E.R. is a consultant for Ception Therapeutics, Merck Research Laboratory, Centocor, Biocryst Pharmaceuticals, and Nycomed.

This work was supported in part by the Thrasher Research Fund NR-0014 (to C.B.), the 2009 American Partnership for Eosinophilic Disorders HOPE research grant (to C.B.), Public Health Service Grant P30 DK0789392 (to C.B.), National Institutes of Health Grants AI079874-01 (to C.B.), AI070235, AI45898, and DK076893 (to M.E.R.), the Food Allergy and Anaphylaxis Network (to M.E.R.), the Campaign Urging Research for Eosinophil Disorders, the Buckeye Foundation (to M.E.R.), and the Food Allergy Project (to M.E.R).

The online version of this article contains supplemental material.

Abbreviations used in this paper:

     
  • AD

    atopic dermatitis

  •  
  • DN

    dominant-negative

  •  
  • EDC

    epidermal differentiation complex

  •  
  • EE

    eosinophilic esophagitis

  •  
  • EV

    empty vector

  •  
  • GERD

    gastroesophageal reflux disease

  •  
  • hpf

    high-power field

  •  
  • NL

    normal

  •  
  • ns

    nonsignificant

  •  
  • OR

    odds ratio

  •  
  • SPRR

    small proline-rich protein.

1
Furuta
G. T.
,
Liacouras
C. A.
,
Collins
M. H.
,
Gupta
S. K.
,
Justinich
C.
,
Putnam
P. E.
,
Bonis
P.
,
Hassall
E.
,
Straumann
A.
,
Rothenberg
M. E.
First International Gastrointestinal Eosinophil Research Symposium (FIGERS) Subcommittees
.
2007
.
Eosinophilic esophagitis in children and adults: a systematic review and consensus recommendations for diagnosis and treatment.
Gastroenterology
133
:
1342
1363
.
2
Molina-Infante
J.
,
Ferrando-Lamana
L.
,
Mateos-Rodríguez
J. M.
,
Pérez-Gallardo
B.
,
Prieto-Bermejo
A. B.
.
2008
.
Overlap of reflux and eosinophilic esophagitis in two patients requiring different therapies: a review of the literature.
World J. Gastroenterol.
14
:
1463
1466
.
3
Lierse
W.
1992
.
The physiology and pathology of the esophagus.
Eur. J. Pediatr. Surg.
2
:
323
326
.
4
De Benedetto
A.
,
Qualia
C. M.
,
Baroody
F. M.
,
Beck
L. A.
.
2008
.
Filaggrin expression in oral, nasal, and esophageal mucosa.
J. Invest. Dermatol.
128
:
1594
1597
.
5
Crish
J. F.
,
Bone
F.
,
Banks
E. B.
,
Eckert
R. L.
.
2002
.
The human involucrin gene contains spatially distinct regulatory elements that regulate expression during early versus late epidermal differentiation.
Oncogene
21
:
738
747
.
6
Crish
J. F.
,
Gopalakrishnan
R.
,
Bone
F.
,
Gilliam
A. C.
,
Eckert
R. L.
.
2006
.
The distal and proximal regulatory regions of the involucrin gene promoter have distinct functions and are required for in vivo involucrin expression.
J. Invest. Dermatol.
126
:
305
314
.
7
Crish
J. F.
,
Howard
J. M.
,
Zaim
T. M.
,
Murthy
S.
,
Eckert
R. L.
.
1993
.
Tissue-specific and differentiation-appropriate expression of the human involucrin gene in transgenic mice: an abnormal epidermal phenotype.
Differentiation
53
:
191
200
.
8
Kobayashi
M.
,
Naomoto
Y.
,
Nobuhisa
T.
,
Okawa
T.
,
Takaoka
M.
,
Shirakawa
Y.
,
Yamatsuji
T.
,
Matsuoka
J.
,
Mizushima
T.
,
Matsuura
H.
, et al
.
2006
.
Heparanase regulates esophageal keratinocyte differentiation through nuclear translocation and heparan sulfate cleavage.
Differentiation
74
:
235
243
.
9
Marenholz
I.
,
Zirra
M.
,
Fischer
D. F.
,
Backendorf
C.
,
Ziegler
A.
,
Mischke
D.
.
2001
.
Identification of human epidermal differentiation complex (EDC)-encoded genes by subtractive hybridization of entire YACs to a gridded keratinocyte cDNA library.
Genome Res.
11
:
341
355
.
10
Proksch
E.
,
Fölster-Holst
R.
,
Jensen
J. M.
.
2006
.
Skin barrier function, epidermal proliferation and differentiation in eczema.
J. Dermatol. Sci.
43
:
159
169
.
11
Mischke
D.
1998
.
The complexity of gene families involved in epithelial differentiation. Keratin genes and the epidermal differentiation complex.
Subcell. Biochem.
31
:
71
104
.
12
Marenholz
I.
,
Volz
A.
,
Ziegler
A.
,
Davies
A.
,
Ragoussis
I.
,
Korge
B. P.
,
Mischke
D.
.
1996
.
Genetic analysis of the epidermal differentiation complex (EDC) on human chromosome 1q21: chromosomal orientation, new markers, and a 6-Mb YAC contig.
Genomics
37
:
295
302
.
13
Régnier
M.
,
Caron
D.
,
Reichert
U.
,
Schaefer
H.
.
1992
.
Reconstructed human epidermis: a model to study in vitro the barrier function of the skin.
Skin Pharmacol.
5
:
49
56
.
14
Poumay
Y.
,
Dupont
F.
,
Marcoux
S.
,
Leclercq-Smekens
M.
,
Hérin
M.
,
Coquette
A.
.
2004
.
A simple reconstructed human epidermis: preparation of the culture model and utilization in in vitro studies.
Arch. Dermatol. Res.
296
:
203
211
.
15
Presland
R. B.
,
Coulombe
P. A.
,
Eckert
R. L.
,
Mao-Qiang
M.
,
Feingold
K. R.
,
Elias
P. M.
.
2004
.
Barrier function in transgenic mice overexpressing K16, involucrin, and filaggrin in the suprabasal epidermis.
J. Invest. Dermatol.
123
:
603
606
.
16
Vickery
B. P.
2007
.
Skin barrier function in atopic dermatitis.
Curr. Opin. Pediatr.
19
:
89
93
.
17
Morar
N.
,
Cookson
W. O.
,
Harper
J. I.
,
Moffatt
M. F.
.
2007
.
Filaggrin mutations in children with severe atopic dermatitis.
J. Invest. Dermatol.
127
:
1667
1672
.
18
Hanifin
J. M.
2008
.
Filaggrin mutations and allergic contact sensitization.
J. Invest. Dermatol.
128
:
1362
1364
.
19
Novak
N.
,
Baurecht
H.
,
Schäfer
T.
,
Rodriguez
E.
,
Wagenpfeil
S.
,
Klopp
N.
,
Heinrich
J.
,
Behrendt
H.
,
Ring
J.
,
Wichmann
E.
, et al
.
2008
.
Loss-of-function mutations in the filaggrin gene and allergic contact sensitization to nickel.
J. Invest. Dermatol.
128
:
1430
1435
.
20
Weidinger
S.
,
O'Sullivan
M.
,
Illig
T.
,
Baurecht
H.
,
Depner
M.
,
Rodriguez
E.
,
Ruether
A.
,
Klopp
N.
,
Vogelberg
C.
,
Weiland
S. K.
, et al
.
2008
.
Filaggrin mutations, atopic eczema, hay fever, and asthma in children.
J. Allergy Clin. Immunol.
121
:
1203
1209
.
21
Fallon
P. G.
,
Sasaki
T.
,
Sandilands
A.
,
Campbell
L. E.
,
Saunders
S. P.
,
Mangan
N. E.
,
Callanan
J. J.
,
Kawasaki
H.
,
Shiohama
A.
,
Kubo
A.
, et al
.
2009
.
A homozygous frameshift mutation in the mouse Flg gene facilitates enhanced percutaneous allergen priming.
Nat. Genet.
41
:
602
608
.
22
Marenholz
I.
,
Kerscher
T.
,
Bauerfeind
A.
,
Esparza-Gordillo
J.
,
Nickel
R.
,
Keil
T.
,
Lau
S.
,
Rohde
K.
,
Wahn
U.
,
Lee
Y. A.
.
2009
.
An interaction between filaggrin mutations and early food sensitization improves the prediction of childhood asthma.
J. Allergy Clin. Immunol.
123
:
911
916
.
23
Stemmler
S.
,
Nothnagel
M.
,
Parwez
Q.
,
Petrasch-Parwez
E.
,
Epplen
J. T.
,
Hoffjan
S.
.
2009
.
Variation in genes of the epidermal differentiation complex in German atopic dermatitis patients.
Int. J. Immunogenet.
36
:
217
222
.
24
Barker
J. N.
,
Palmer
C. N.
,
Zhao
Y.
,
Liao
H.
,
Hull
P. R.
,
Lee
S. P.
,
Allen
M. H.
,
Meggitt
S. J.
,
Reynolds
N. J.
,
Trembath
R. C.
,
McLean
W. H.
.
2007
.
Null mutations in the filaggrin gene (FLG) determine major susceptibility to early-onset atopic dermatitis that persists into adulthood.
J. Invest. Dermatol.
127
:
564
567
.
25
Blanchard
C.
,
Mingler
M. K.
,
Vicario
M.
,
Abonia
J. P.
,
Wu
Y. Y.
,
Lu
T. X.
,
Collins
M. H.
,
Putnam
P. E.
,
Wells
S. I.
,
Rothenberg
M. E.
.
2007
.
IL-13 involvement in eosinophilic esophagitis: transcriptome analysis and reversibility with glucocorticoids.
J. Allergy Clin. Immunol.
120
:
1292
1300
.
26
Akei
H. S.
,
Brandt
E. B.
,
Mishra
A.
,
Strait
R. T.
,
Finkelman
F. D.
,
Warrier
M. R.
,
Hershey
G. K.
,
Blanchard
C.
,
Rothenberg
M. E.
.
2006
.
Epicutaneous aeroallergen exposure induces systemic TH2 immunity that predisposes to allergic nasal responses.
J. Allergy Clin. Immunol.
118
:
62
69
.
27
Akei
H. S.
,
Mishra
A.
,
Blanchard
C.
,
Rothenberg
M. E.
.
2005
.
Epicutaneous antigen exposure primes for experimental eosinophilic esophagitis in mice.
Gastroenterology
129
:
985
994
.
28
Mishra
A.
,
Rothenberg
M. E.
.
2003
.
Intratracheal IL-13 induces eosinophilic esophagitis by an IL-5, eotaxin-1, and STAT6-dependent mechanism.
Gastroenterology
125
:
1419
1427
.
29
Kim
B. E.
,
Leung
D. Y.
,
Boguniewicz
M.
,
Howell
M. D.
.
2008
.
Loricrin and involucrin expression is down-regulated by Th2 cytokines through STAT-6.
Clin. Immunol.
126
:
332
337
.
30
Blanchard
C.
,
Wang
N.
,
Stringer
K. F.
,
Mishra
A.
,
Fulkerson
P. C.
,
Abonia
J. P.
,
Jameson
S. C.
,
Kirby
C.
,
Konikoff
M. R.
,
Collins
M. H.
, et al
.
2006
.
Eotaxin-3 and a uniquely conserved gene-expression profile in eosinophilic esophagitis.
J. Clin. Invest.
116
:
536
547
.
31
Dennis
G.
 Jr.
,
Sherman
B. T.
,
Hosack
D. A.
,
Yang
J.
,
Gao
W.
,
Lane
H. C.
,
Lempicki
R. A.
.
2003
.
DAVID: Database for Annotation, Visualization, and Integrated Discovery.
Genome Biol.
4
:
P3
.
32
Hosack
D. A.
,
Dennis
G.
 Jr.
,
Sherman
B. T.
,
Lane
H. C.
,
Lempicki
R. A.
.
2003
.
Identifying biological themes within lists of genes with EASE.
Genome Biol.
4
:
R70
.
33
Blanchard
C.
,
Durual
S.
,
Estienne
M.
,
Emami
S.
,
Vasseur
S.
,
Cuber
J. C.
.
2005
.
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.
37
:
2559
2573
.
34
Enomoto
H.
,
Hirata
K.
,
Otsuka
K.
,
Kawai
T.
,
Takahashi
T.
,
Hirota
T.
,
Suzuki
Y.
,
Tamari
M.
,
Otsuka
F.
,
Fujieda
S.
, et al
.
2008
.
Filaggrin null mutations are associated with atopic dermatitis and elevated levels of IgE in the Japanese population: a family and case-control study.
J. Hum. Genet.
53
:
615
621
.
35
Smith
F. J.
,
Irvine
A. D.
,
Terron-Kwiatkowski
A.
,
Sandilands
A.
,
Campbell
L. E.
,
Zhao
Y.
,
Liao
H.
,
Evans
A. T.
,
Goudie
D. R.
,
Lewis-Jones
S.
, et al
.
2006
.
Loss-of-function mutations in the gene encoding filaggrin cause ichthyosis vulgaris.
Nat. Genet.
38
:
337
342
.
36
Sharma
M.
,
Mehla
K.
,
Batra
J.
,
Ghosh
B.
.
2007
.
Association of a chromosome 1q21 locus in close proximity to a late cornified envelope-like proline-rich 1 (LELP1) gene with total serum IgE levels.
J. Hum. Genet.
52
:
378
383
.
37
Noel
R. J.
,
Putnam
P. E.
,
Collins
M. H.
,
Assa’ad
A. H.
,
Guajardo
J. R.
,
Jameson
S. C.
,
Rothenberg
M. E.
.
2004
.
Clinical and immunopathologic effects of swallowed fluticasone for eosinophilic esophagitis.
Clin. Gastroenterol. Hepatol.
2
:
568
575
.
38
Zuo
L.
,
Fulkerson
P. C.
,
Blanchard
C.
,
Rothenberg
M. E.
.
2007
.
Inducible expression of the IL-13 transgene in the lung results in IL-13 accumulation, eosinophilia, and tissue remodeling in the esophagus.
J Allergy Clin Immunol.
119
:
S152
.
39
Carlsson
R.
,
Fändriks
L.
,
Jönsson
C.
,
Lundell
L.
,
Orlando
R. C.
.
1999
.
Is the esophageal squamous epithelial barrier function impaired in patients with gastroesophageal reflux disease?
Scand. J. Gastroenterol.
34
:
454
458
.
40
Petering
H.
,
Kluthe
C.
,
Dulkys
Y.
,
Kiehl
P.
,
Ponath
P. D.
,
Kapp
A.
,
Elsner
J.
.
2001
.
Characterization of the CC chemokine receptor 3 on human keratinocytes.
J. Invest. Dermatol.
116
:
549
555
.
41
Amerio
P.
,
Frezzolini
A.
,
Feliciani
C.
,
Verdolini
R.
,
Teofoli
P.
,
De Pità
O.
,
Puddu
P.
.
2003
.
Eotaxins and CCR3 receptor in inflammatory and allergic skin diseases: therapeutical implications.
Curr. Drug Targets Inflamm. Allergy
2
:
81
94
.
42
Ellyard
J. I.
,
Simson
L.
,
Bezos
A.
,
Johnston
K.
,
Freeman
C.
,
Parish
C. R.
.
2007
.
Eotaxin selectively binds heparin. An interaction that protects eotaxin from proteolysis and potentiates chemotactic activity in vivo.
J. Biol. Chem.
282
:
15238
15247
.
43
Yuan
Q.
,
Campanella
G. S.
,
Colvin
R. A.
,
Hamilos
D. L.
,
Jones
K. J.
,
Mathew
A.
,
Means
T. K.
,
Luster
A. D.
.
2006
.
Membrane-bound eotaxin-3 mediates eosinophil transepithelial migration in IL-4-stimulated epithelial cells.
Eur. J. Immunol.
36
:
2700
2714
.
44
Konikoff
M. R.
,
Blanchard
C.
,
Kirby
C.
,
Buckmeier
B. K.
,
Cohen
M. B.
,
Heubi
J. E.
,
Putnam
P. E.
,
Rothenberg
M. E.
.
2006
.
Potential of blood eosinophils, eosinophil-derived neurotoxin, and eotaxin-3 as biomarkers of eosinophilic esophagitis.
Clin. Gastroenterol. Hepatol.
4
:
1328
1336
.
45
Palmer
C. N.
,
Irvine
A. D.
,
Terron-Kwiatkowski
A.
,
Zhao
Y.
,
Liao
H.
,
Lee
S. P.
,
Goudie
D. R.
,
Sandilands
A.
,
Campbell
L. E.
,
Smith
F. J.
, et al
.
2006
.
Common loss-of-function variants of the epidermal barrier protein filaggrin are a major predisposing factor for atopic dermatitis.
Nat. Genet.
38
:
441
446
.