Share this article on:

Angiogenic Remodeling in Pediatric EoE Is Associated With Increased Levels of VEGF-A, Angiogenin, IL-8, and Activation of the TNF-α–NFκB Pathway

Persad, Rabindranath*; Huynh, Hien Q.*; Hao, Li*; Ha, Jacqueline R.*; Sergi, Consolato; Srivastava, Rahul*; Persad, Sujata*

Journal of Pediatric Gastroenterology & Nutrition: September 2012 - Volume 55 - Issue 3 - p 251–260
doi: 10.1097/MPG.0b013e31824b6391

Objectives: Eosinophilic esophagitis (EoE) is a clinicopathologic diagnosis characterized by inflammation and infiltration of eosinophils at the esophageal mucosa. The underlying etiology of EoE remains elusive. Inflammatory diseases, such as asthma, are associated with structural remodeling of the airways, which includes angiogenesis. The aims of this study were to determine the angiogenic profile of esophageal mucosa in children presenting with EoE and to evaluate the putative mechanism(s) underlying the early inflammatory angiogenic response observed in EoE.

Methods: Endoscopically obtained biopsy samples from 18 EoE and 18 control pediatric patients were analyzed for angiogenic markers (CD31, von Willebrand factor, vascular cell adhesion molecule-1) and tissue levels of angiogenic factors (vascular endothelial growth factor [VEGF]-A, VEGF-R2, angiogenin and interleukin [IL]-8). Expression levels of angiogenic factors and markers in EoE and control samples were characterized by immunofluorescence analysis and quantitative reverse transcriptase-polymerase chain reaction. Vascular density of biopsy samples was evaluated by immunofluorescence analysis.

Results: Samples from patients with EoE exhibited higher levels of von Willebrand factor, CD31, and vascular cell adhesion molecule-1, which is suggestive of neovascularization and an activated endothelium. Moreover, EoE biopsies showed greater levels of the angiogenesis promoters VEGFA, angiogenin, and IL-8. Interestingly, there were greater cellular levels of tumor necrosis factor-α in EoE samples compared with controls. Furthermore, there were higher nuclear levels of p50 and p65 subunits of NFκB and lower cellular levels of the inhibitor of NFκB, IκB-α, in EoE samples compared with controls.

Conclusions: We demonstrate increased angiogenesis in the esophageal mucosa of pediatric patients with EoE. The data also provided evidence that the angiogenic factors VEGF-A, angiogenin, and IL-8 were prominently involved in promoting angiogenic remodeling.

*Department of Paediatrics

Department of Laboratory Medicine and Pathology, University of Alberta, Edmonton, Alberta, Canada.

Address correspondence and reprint requests to Dr Sujata Persad, Department of Paediatrics, University of Alberta, 3020 KATZ Group Centre for Pharmacy and Health Research, Edmonton, AB, Canada T6G 2E1 (e-mail:

Received 22 February, 2011

Accepted 1 August, 2011

Drs Persad and Huynh are co-first authors.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Website (

This project was funded by the Stollery Children's Hospital Research Foundation, Edmonton.

The authors report no conflicts of interest.

Gastroesophageal reflux disease (GERD) and eosinophilic esophagitis (EoE) are 2 of the most common nonneoplastic esophageal diseases in both children and adults. Passive flow of gastric content back into the esophagus characterizes gastroesophageal reflux. Typical GERD symptoms include heartburn with or without endoscopic/histologic evidence of tissue damage (1); however, the relevance of these symptoms remains a source of debate. EoE is a clinicopathologic entity, combining clinical history findings (food impaction, dysphagia, feeding intolerance failure to thrive, classic GERD symptoms), with histologic data (≥15 eosinophils/high-power field), and clinicotherapeutic data (persistence of tissue eosinophilia despite proton pump inhibitor [PPI] use) (2,3). In addition, clinicodiagnostic data must exclude other diseases with overlapping features, including GERD, Crohn disease, and eosinophilic gastroenteritis (2,3). The prevalence of EoE has notably increased in the last 5 years, but its etiology has not been elucidated. Hypotheses are centered on an immunologic or immunoallergic basis (4) and patients with EoE have seasonal variations in symptoms, providing clinical evidence to support a role for aeroallergen-driven eosinophil-associated responses in the esophagus (5).

A pathologic immunoallergic response to food allergens has also been hypothesized, which involves an allergen-induced, T-helper 2 (TH2)-cytokine–dependent, interleukin (IL)-5–mediated infiltration of eosinophils to the esophageal mucosa (2,6). The majority of patients with EoE are atopic, based on the coexistence of atopic dermatitis, allergic rhinitis, and/or asthma (2,6). This association has led to speculation that the tissue remodeling observed in EoE may be similar to that described in asthmatic patients (7).

Inflammatory diseases, such as asthma, have been reported to undergo varying degrees of structural changes termed airways remodeling, which leads to subepithelial fibrosis, smooth muscle hypertrophy, and angiogenesis (7). Moreover, structural remodeling of the esophagus in EoE is underscored by gross and microscopic changes, including fibrosis of the lamina propria (8). In asthma, angiogenesis plays a key role in remodeling, resulting in an increased number of conduits for inflammation. In fact, the role of angiogenesis is recognized as a potent regulator of inflammation in several disease states (9,10); however, little is presently known with regard to the role of angiogenic remodeling in EoE. Recently, vascular density of the lamina propria was found to be increased in patients with EoE relative to patients with GERD and/or to healthy individuals; this has been associated with endothelial activation and expression of vascular cell adhesion molecule 1 (VCAM-1) (11).

Angiogenesis is a multistep process in which endothelial cells, upon receiving a stimulatory signal, become activated to undergo proliferation and migration (12). This process is regulated by the balance of proangiogenic and inhibitory factors, commonly referred to as the angiogenic switch (13). Angiogenic stimulators include vascular endothelial growth factor (VEGF), angiogenin, and IL-8 (14). Several lines of evidence have indicated that VEGF is one of the most potent proangiogenic factors for both physiologic and pathologic angiogenic settings (15) and functions by stimulating endothelial cell migration and proliferation (16,17). There are four different isoforms of human VEGF (VEGF-A through -D), with VEGF-A being the most angiogenic under physiologic and pathologic conditions (18). Present evidence suggests VEGF-A is a mediator of blood vessels and extravascular tissue remodeling and inflammation (19). Similarly, angiogenins, which are another group of angiogenic stimulators, have been identified to work independently and in combination with VEGF-A to regulate neovascularization (20,21). Angiogenins interact with endothelial and smooth muscle cells to induce a broad range of biologic responses, including cell proliferation, migration, and tubule formation (22,23). Endothelial cell proliferation and survival are also enhanced by IL-8, a CXC chemokine that is produced by epithelial cells and macrophages (24–26).

To date, there are no reports of the mechanism of angiogenic remodeling in EoE. Elucidating this mechanism will have significant therapeutic and prognostic relevance for patients with EoE. We speculated that increased angiogenesis is integral to the early inflammation observed in the pediatric presentation of EoE. Greater vascular density would be integral to maintaining the inflammatory process.

The aim of the present study was to determine whether angiogenic changes occur in the esophageal mucosa of children presenting with EoE. Moreover, the present study evaluated putative mechanisms that may play a role in the pathogenesis of eosinophilic granulocyte recruitment in the early inflammatory response observed in EoE.

Back to Top | Article Outline


Sample Selection

This project received approval from the health research ethics board at the University of Alberta and Alberta Health Services, Edmonton. Esophageal samples from patients with EoE were selected from our EoE clinicopathologic database (presently includes >100 children), who presented with childhood onset of EoE, before 18 years of age. All of the included patients met diagnostic criteria for the absence of pathologic acid reflux, based on 24-hour pH probe criteria or failure to respond to antireflux therapy with PPI 4 weeks before endoscopy. Samples were selected for inclusion when there was endoscopic evidence of EoE despite aggressive PPI therapy. This included both macroscopic presence of linear furrows, white exudates, and mucosal edema. There was no anatomical division into upper versus lower esophagus ± mid esophagus. All of the biopsies were reviewed by board-certified pediatric pathologists, and eosinophils were manually quantified.

Control samples were selected from our endoscopic database of a patient population that underwent endoscopy for recurrent abdominal pain or suspected GERD. In all of the control samples, either a macroscopically normal-appearing esophagus or loss of vascular pattern and/or erosions in the lower esophagus was reported. Less than 5 eosinophils per high-power field were present in all of the biopsy samples. Patients with other inflammatory conditions including celiac disease, inflammatory bowel diseases, polyposis syndromes, or hypereosinophilic syndrome were excluded. Random biopsy samples were obtained from 18 pediatric patients in the EoE clinic database and 18 control patients from the pediatric endoscopy database meeting the above set of criteria.

Back to Top | Article Outline

Immunofluorescence Analysis

Tissue sections (∼4-μm thick) were dewaxed, hydrated, and blocked for nonspecific binding with 10% bovine serum albumin. Vascular density was evaluated by immunofluorescence (IF) analysis with antibodies to von Willebrand factor (vWF) and CD31. Biopsy samples were also stained for VCAM-1 to detect the activation state of the vascular endothelium. VCAM-1 is a member of the immunoglobulin (Ig) superfamily of proteins found on endothelium (27). The presence of eosinophils in the esophageal tissue of each group was demonstrated by staining for major basic protein (MBP), the degranulation product of eosinophils that plays a role in the release of inflammatory mediators (28).

Tissue sections were first incubated with the primary antibody (anti-vWF [Dako, Carpinteria, CA]; anti CD31, anti-VCAM-1 [R&D Systems, Minneapolis, MN]; or anti-MBP [BD Biosciences, San Jose, CA]) and then incubated with the appropriate Alexafluor-488- or Alexafluor-555–conjugated secondary fluorescent antibody (Invitrogen, Carlsbad, CA). Negative controls included replacing the primary antibody with normal rabbit serum. Immunostained slides were quantified by scanning tissue sections under low power (100×) to detect hot spots, and the number of cells or vessel density were counted at higher magnification (400×) in 3 to 5 fields within these hot spots. Data were evaluated by computer-assisted quantification by 2 independent observers blinded to the clinical parameters of the samples. Results were expressed as number of cells or number of vessels per high-power field.

Parallel tissue sections were immunostained for VEGF-A (Santa Cruz Biotechnology, Santa Cruz, CA), VEGF receptor-1 and -2 (VEGF-R1 and VEGF-R2) (Cell Signaling Danvers, MA), angiogenin (Cedarlane), IL-8 (Invitrogen), tumor necrosis factor-α (TNF-α) (Santa Cruz), nuclear factor κB (NFκB) subunits p65 and p50 (Santa Cruz), and IκB-α (Cell Signaling) using the respective primary antibody and then with the appropriate Alexafluor-488–conjugated secondary fluorescent antibody. Expressions of the individual proteins were determined with microscopy and staining intensity was quantified by Metamorph (Molecular Devices, Sunnyvale, CA).

Back to Top | Article Outline

Statistical Analysis for IF Analysis

It was determined that a sample size of 18 biopsies from each group was required for the present study. Assuming an α of 0.007 (Bonferroni corrected), 1-way analysis of variance with 18 samples per group had 95% power to detect a difference in the group vascular density means of 50 and 22, assuming Smax = 24. Differences and variability were based on the previous data of Aceves et al (11). Sigma Plot software (Systat Software, San Jose, CA) was used for statistical analysis including calculation of the mean and standard error of EoE and control groups. Two-tailed unpaired Student t test was used for comparison of means. Probability values of <0.01 were considered statistically significant.

Back to Top | Article Outline

Quantitative Real-time Polymerase Chain Reaction

Total RNA was isolated from esophageal tissue from control and patients with EoE, using the RNeasy Mini Kit (Qiagen, Valencia, CA) and treated with DNase I (Invitrogen). Isolated RNA was used for first-strand cDNA synthesis, using oligo (dT) (Invitrogen) and superscript II reverse transcriptase (Invitrogen). Real-time quantification of IL-8, angiogenin, CD31, VCAM-1, VEGF-R2, and VEGF-A was assessed using Power SYBR Green PCR Master Mix (P/N: 4367659; Applied Biosystems, Foster City, CA). GAPDH served as the endogenous control. Samples were amplified with a precycling hold at 95°C for 15 seconds, annealing and extension at 60°C for 1 minute. Primers used are as follows: CD 31-sense: 5′-CTGCCTTCCTTCGGGTTGCA-3′, antisense: 5′-CAAGGACTCACCTTCCACCAACAG-3′; VCAM-1-sense: 5′-GAAAAGTTC TTGTTTGCCGAGC-3′, antisense: 5′-GCAGCTTTGTGGATGGATTCAC-3′; VEGF-A-sense: 5′-GAAAAGTTCTTGTTTGCCGAGC-3′, antisense: 5′-TCACCGCCT CGGCTTGTCACA-3′; VEGF-R2-sense: 5′-GAGTACCCTGATGAGATCGAG-3′, antisense: 5′-TGGACAAGTAGCCTGTCTTCAGTT-3′; angiogenin-sense: 5′-CCTGGGCGTTTTGTTGTTGG-3′, antisense: 5′-TGTGGCTCGGTACTGGCATG-3′; IL-8-sense: 5′-AGGTGCAGTTTTGCCAAGGA-3′, antisense: 5′-TTTCTGTGTT GGCGCAGTGT-3′; GAPDH-sense: 5′-TCAACGACCACTTTGTCAAGCTCA-3′, antisense: 5′-GCTGGTGGTCCAGGGGTCTTACT-3′.

Each experiment was independently performed using the Rotor-Gene-3000 (Montreal Biotech, Montreal, Canada). Each data point was an average of 6 experiments done in triplicate using tissue from different donors. Data were analyzed using Rotor-Gene-6 software (Qiagen, Valencia, CA). Gene expression was determined using the relative standard curve method normalized to GAPDH expression. Results were expressed as a fold change in the target mRNA levels. Sigma Plot software was used for statistical analysis including calculation of the mean and standard error of EoE and control groups. Student t test and Mann-Whitney rank sum test were used for comparison of means. Probability values of <0.01 were considered statistically significant.

Back to Top | Article Outline


Patients’ Characteristics

The mean age of patients with EoE (n = 18) and control (n = 18) was 12 and 9.9 years, respectively (Table 1). Seventy-eight percent of the children with EoE were male and 22% were female. In the control group, 56% of the children were boys and 44% were girls. Eighty-nine percent of the children with EoE complained of dysphagia, whereas 10% of the normal children complained of dysphagia. Fourteen of the 18 patients with EoE were atopic (78%), whereas 1 of the 18 children in the control group was atopic. Table 1 details the frequency of other symptoms in the 2 groups.

Back to Top | Article Outline

Endoscopic Characteristics

None of the patients with EoE had a normal endoscopy of the esophagus. Endoscopic features of patients with EoE are shown in Table 1. Features included loss of vascular pattern (72%), vertical furrows (56%), concentric rings (17%), and white exudates (30%). All of the patients with EoE had a normal endoscopy for stomach and small intestines. All of the controls had a normal endoscopy of the esophagus. Five control (GERD) patients showed minor erosive changes of the distal esophagus.

Back to Top | Article Outline

Tissue Eosinophilia

The mean number of esophageal eosinophils was 65 eosinophils (Eos)/hpf (range 30–100) in patients with EoE compared with 3 Eos/hpf in patients with GERD (range 0–5), which served as our control group. The patients had normal gastric and duodenal biopsies. Original assessment of eosinophil count was done with H&E stain by pathologists within the Department of Laboratory Medicine and Pathology, University of Alberta. To further demonstrate the presence of eosinophils, esophageal tissue of each group was stained for MBP. Figure 1 indicates that there was significantly higher density of MBP staining in the biopsy samples from the patients with EoE relative to control tissue.

Back to Top | Article Outline

Tissue Vascularity and Neovascularization

Vascular density was evaluated by staining tissue sections with anti-vWF and CD31 antibody. vWF is produced by endothelial cells and is widely used in immunohistochemistry to visualize blood vessels (29). CD31 is an endothelial adhesion molecule involved in the process of angiogenesis (30). Anti-vWF antibody was used for identification of all vessels (both mature and newly formed), and an antibody to CD31 was used to identify newly formed vessels. The use of anti-vWF and anti-CD31 ensures measurement of active vascular remodeling rather than a static vessel count. Our data demonstrated that there were significantly greater numbers of blood vessels in the subepithelial region of the EoE esophageal tissue samples compared with control samples (Fig. 2A and B). Furthermore, patients with EoE had an average of 75 vessels per high-power field relative to the control group with 13 vessels per high-power field. Similar results were observed when biopsy samples were immunostained for vWF (EoE: 79 vessels per high-power field; control: 16 vessels per high-power field) (Fig. 2C). Collectively, these results suggested significantly increased levels of neovascularization within the esophageal mucosa in EoE.

Adhesion of eosinophils to VCAM-1 and the resulting migration of eosinophils were proposed to be crucial for the development of selective inflammatory sites in various eosinophil-driven disease states (27). As shown in Figure 2D and E, the number of VCAM-1–positive vessels within the subepithelia was significantly greater in the EoE tissue (mean of 41 vessels per high-power field) compared with the control (mean of 6 vessels per high-power field). This is indicative of a significantly greater activated endothelial component in the subepithelia of the EoE tissue compared with control.

Back to Top | Article Outline

Tissue Expression of Stimulators of Angiogenesis: VEGF-A, Angiogenin, and IL-8

We next compared the levels of VEGF-A, VEGF-R1, and VEGF-R2 in EoE and control esophageal biopsy tissue. VEGF-A, when bound to its receptor, VEGF-R2, is a potent promoter of tissue angiogenesis (31). Upon IF analysis, we observed that there was a significantly greater number of cells that stained positive for both VEGF-A and VEGF-R2 in the EoE samples (Fig. 3A and B). For VEGF-A, there was an average of 169 positively stained cells per high-power field in EoE tissue compared with a mean of 32 positively stained cells per high-power field in control. Similarly, for VEGF-R2 there was an average of 173 positively stained cells per high-power field in EoE relative to 28 positively stained cells per high-power field in the control. Because VEGF-A has been shown to bind to both VEGF-R1 and VEGF-R2, we also evaluated the number of VEGF-R1-positive cells. Our results indicated that the number of VEGF-R1-positive cells was similar (mean of 7/hpf) for both EoE and control mucosal tissue (data not shown).

In addition to greater positive staining of VEGF-A in patients with EoE, IF analysis for angiogenin showed an approximately 4-fold greater number of positively stained cells in EoE tissue (mean of 191 cells/high-power field) compared with control (mean of 49 cells/high-power field) (Fig. 3C).

IL-8 is known to support angiogenesis by enhancing endothelial cell proliferation and survival (24–26,32), and preventing apoptosis of endothelial cells expressing its receptors. Figure 3D demonstrates that there was a 9-fold greater number of IL-8–positive cells in EoE biopsy tissue (mean of 176 cells/hpf) compared with control (mean of 20 cells/hpf).

Back to Top | Article Outline

Tissue mRNA Expression of CD31, VCAM-1, VEGF-A, Angiogenin, and IL-8

We carried out quantitative real-time polymerase chain reaction using total RNA extracts from our EoE and control mucosal biopsy tissue. Figure 4 represents mRNA expression levels of the endothelial and angiogenic markers. Expressions of CD31, VCAM-1, VEGF-A, angiogenin, and IL-8 were consistently greater in EoE samples compared with controls. The changes in angiogenin mRNA levels were more profound relative to that of VEGF-A mRNA levels. Moreover, our data also indicated that increased mRNA levels of these markers were directly correlated with eosinophil numbers in the tissue biopsies. As seen in Figure 4 A–E, there was significantly greater mRNA expression of the endothelial and angiogenic markers in biopsy tissues with >15 Eos compared with the corresponding mRNA levels in tissue with <15 Eos.

Back to Top | Article Outline

Tissue Expression of TNF-α, NFκB, and IκB-α

In an effort to understand the putative mechanism(s) regulating the proangiogenic activity in the EoE esophageal mucosa, we compared the protein expression level of TNF-α. TNF-α is a proinflammatory cytokine that plays a role in inducing a number of inflammatory conditions (including asthma) as well as regulating angiogenesis (33). Our data indicated that there was a 7-fold greater number of TNF-α–positive cells in the EoE biopsies (mean of 184/hpf) compared with the control (mean of 24/hpf) (Fig. 5A).

One of the major downstream transcriptional effector pathways of TNF-α signaling is the NFκB pathway (34). To evaluate whether NFκB played a role downstream of TNF-α in EoE, we determined the expression and localization of p50/p65 heterodimer of NFκB in EoE compared with control. Our data indicated that although there was little difference in the cellular levels of p50 and p65 subunits between control and EoE mucosal samples, patients with EoE expressed significantly higher levels of both p65 and p50 in the nuclei of the mucosal cells compared with controls (Fig. 5B and C). In contrast, there were significantly less immunostaining for inhibitor of NFκB activity, IκB-α, in EoE mucosal samples (mean of 46 cells/hpf) compared with control (mean of 189 cells/hpf) (Fig. 5D).

Back to Top | Article Outline


It is important that novel pathogenic mechanisms be investigated in pediatric patients with EoE because it is seems likely that the inflammatory process beginning in childhood can lead to structural changes in the esophagus that become refractory to anti-inflammatory strategies. Certainly the need for esophageal dilatation has been observed in adult patients with EoE, and this group has been identified as having a high risk for endoscopic complications (35). The same is not true for pediatric patients, and early intervention may halt disease progression. There has been a dearth of basic science investigations into the pathogenic mechanisms of EoE, and hence the development of novel therapies has been wanting. The present study brings new and important insights into pathogenesis of early-onset EoE.

Here, we demonstrated significant angiogenic remodeling in the esophageal mucosa of pediatric patients with EoE. The increase in vascular density, demonstrated by the greater number of vWF- and CD31-positive stained vessels in the esophageal subepithelium in EoE biopsies, strongly suggested neovascularization due to increased angiogenic activity. CD31 is an endothelial intercellular adhesion molecule that is involved in the initial production of endothelial cell-to-cell adhesions, a valuable step in the formation of new vasculature (angiogenesis); however, existing cellular adhesions do not use CD31 to maintain the integrity of the blood vessel structure (30). Therefore, although the greater levels of vWF-stained vessel in the EoE biopsies indicated increased tissue vascularity, the increased number of CD31-positive vessels suggested that this increased vascularity was attributed to augmented neovascularization in the EoE esophageal mucosa.

Neovascularization leads to the expansion of tissue microvasculature (36,37), which allows for an increased influx of inflammatory cells, including eosinophils, and a source of local cytokine production from activated vascular epithelium (38). To this end, VCAM-1 expressed on the surface of endothelial cells serves an important mechanism for eosinophil adhesion to esophageal blood vessels and the subsequent recruitment of eosinophils to esophageal tissue (11). Increased levels of VCAM-1 observed in the EoE esophageal mucosa samples support the involvement of this mechanism in the promotion of EoE. This is further reinforced by the significant localization of the eosinophil degranulation product, MBP, in areas with high expression of VCAM-1. (Supplemental Fig. 1,, shows coimmunostaining of VCAM-1 and MBP. Photomicrographs show colocalization of both proteins in significantly greater proportion in EoE tissues compared with control tissues. Photomicrographs are representative of a sample size of 10 controls and 10 patients with EoE.)

The process of angiogenesis includes endothelial cell proliferation, survival, migration, and capillary tube formation. An imbalance in favor of proangiogenic factors is thought to lead to the abnormal growth of new blood vessels (39). Moreover, significant upregulation of angiogenin immunoreactivity in bronchial asthma has been correlated with the degree of airway vascularity (40,41); however, involvement of angiogenin has not been demonstrated in EoE, and our study was the first to identify angiogenin as a potent promoter of angiogenesis in EoE. Similarly, although IL-8 levels were shown to be upregulated in asthma (42), our study was the first to identify the involvement of IL-8 in EoE. We have also found significantly greater number of cells stained positively for VEGF-A and its receptor VEGF-R2 in EoE compared with the control group; however, levels of VEGF-R1 were not altered in EoE compared with control. Although VEGF-A binds to both VEGF-R1 and -R2 receptors, binding to VEGF-R2 presents its principal signaling pathway for angiogenesis (43). This is due to the fact that the tyrosine kinase activity of VEGF-R1 upon binding to VEGF-A is stimulated weakly compared with that of VEGF-R2. The exact role of VEGF-R1 in endothelial cells, apart from serving as a reservoir for VEGF-A, is disputed. In fact, the sequestration of VEGF-A by VEGF-R1 has been suggested to play a negative regulatory role upon VEGF-R2-mediated induction of angiogenesis and vascular development (43). Furthermore, it has been demonstrated that although angiogenic isoforms of VEGF-A bind VEGF-R2, the antiangiogenic isoforms of VEGF-A bind VEGF-R1 (44). Thus, the increased levels of VEGF-R2 expression observed in EoE in the background of unaltered VEGF-R1 levels are supportive of an increased angiogenic response. It should be clarified that VEGF-R1 does induce a positive angiogenic response when it is bound by the VEGF family member placental growth factor. Although all 3 factors (VEGF-A, IL-8, and angiogenin) are acknowledged as potent promoters of angiogenesis, to the best of our knowledge, their prominent role in promoting neoangiogenesis in EoE has not been reported previously.

Our observations at the protein level were confirmed quantitatively by the mRNA expressions of the endothelial (CD31, VCAM-1) and angiogenic (VEGF-A, angiogenin, and IL-8) markers, which exhibited a parallel pattern of alteration to that of their respective protein levels. This also confirmed the specificity of the alterations and ruled out the possibility of artifactual manifestation. Furthermore, the alterations at the mRNA level suggested that mechanistically, the expression of these proteins was regulated at the transcriptional level.

We investigated the putative involvement of the TNF-α pathway in the regulation of these markers and in promoting angiogenic remodeling in EoE. Moreover, TNF-α is a major proinflammatory cytokine involved in chronic inflammatory diseases of the bones, skin, or bowel. TNF-α is also an important regulator in coordinating inflammation with the onset of angiogenesis, likely via the activation of the NFκB transcription factor family of proteins (34). The binding of TNF-α to its receptor increases the transcription of proinflammatory genes, which includes IL-8. TNF-α also acts as a chemoattractant for neutrophil and eosinophil migration (45), which is partially achieved by TNF-α–mediated increase in the expression of adhesion molecules such as VCAM-1 on endothelial cells (46). Our results demonstrated significantly higher levels of TNF-α in the EoE mucosal biopsies compared with controls. These findings were in agreement with previous studies that reported increased tissue expression of TNF-α in human EoE biopsy samples (47,48). Furthermore, studies done with murine models of allergic inflammation have revealed important roles of TNF-α in eosinophil adhesion to inflamed blood vessels as well as eosinophil recruitment (49). TNF-α is suggested to be at least partially responsible for vascular activation and eosinophil transmigration in the presence of other chemokines (47,48,50).

The NFκB pathway is a major effector of gene expression downstream of the TNF-α signaling pathway (51). The mammalian NFκB family is composed of 5 distinct but structurally related subunits, NFκB1/p50, NFκB2/p52, RelA/p65, RelB, and c-Rel, which can form various homodimeric and heterodimeric combinations, each with a specific signaling characteristic (52). The common form of NFκB is p65/RelAp50. In most cell types, the p65/RelA-p50 heterodimer is sequestered in the cytoplasm by IκB because IκB binding masks the nuclear localization sequence of p65. Upon stimulation of the NFκB pathway, IκB-α is phosphorylated at critical serine residues, resulting in its proteosomal degradation. The p65/RelA-p50 dimer then translocates into the nucleus to bind NFκB-specific enhancer regions and regulates the transcriptional activity of target genes (52). Our results demonstrated a greater expression of p50 and p65 subunits of NFκB in the nuclei of mucosal cells without significant alteration in the total cellular levels of the heterodimers. Furthermore, there was significantly lower expression of IκB-α in the EoE tissue compared with control.

Collectively, these results suggested that EoE is associated with an activated TNF-α-NFκB pathway promoted by increased nuclear localization of the NFκB p50-p65 heterodimer. The increased nuclear localization of the NFκB p50-p65 heterodimer may be a result of the decreased cellular levels of IκB-α, likely due to increased degradation of the latter; however, it should be noted that the alterations in the localization of the p50 and p65 subunits of NFκB were less dramatic than the observed increase in the expression of TNF-α (7-fold). This may indicate that the NFκB is only partially responsible for mediating the TNF-α–induced response and that TNF-α also likely uses other additional signaling/transcriptional pathways to fully exert its effect. It should be pointed out that these findings are correlative at present, and future prospective studies are required to confirm the mechanistic details of these putative regulations.

A limitation of our study was the comparison of EoE to a disease control group that was positive for GERD. The role of acid reflux in etiology and perpetuation of EoE is poorly understood; however, although antacid reflux treatment may improve the symptoms of EoE, there is no evidence to date that it changes EoE histology. Furthermore, we executed rigorous sampling methodology to select EoE samples from patients who represented the extremes for both macroscopic and microscopic features of EoE despite aggressive antacid therapy. The controls in comparison had limited histological evidence for tissue eosinophilia, and consequently we believe disease overlap between EoE and GERD in our control samples was extremely unlikely.

In conclusion, our data demonstrated increased angiogenesis in the subepithelia of the esophageal mucosa in pediatric patients with EoE. This supported the hypothesis that angiogenesis was implicated in the pathogenesis of EoE and that angiogenesis played a role in the esophageal remodeling observed in EoE (13,53). The data also provided evidence that the angiogenic factors VEGF-A, angiogenin, and IL-8 were prominently involved in promoting angiogenic remodeling. Furthermore, we speculate that activation of the TNF-α-NFκB pathway played a role, at least partially, in mediating and triggering the inflammatory response–induced angiogenesis. Prospective studies will help us to further confirm the roles of angiogenin, IL-8, VEGF-A, and TNF-α-NFκB pathway in the pathogenesis of EoE as well as guide the development of novel combination of antiangiogenic and anti-TNF-α treatment strategies that may be beneficial in limiting the progression of the disease.

Back to Top | Article Outline


1. Orlando RC. The integrity of the esophageal mucosa. Balance between offensive and defensive mechanism. Best Pract Res Clin Gastroenterol 2010; 24:873–882.
2. Furuta GT, Liacouras CA, Collins MH, et al. Eosinophilic esophagitis in children and adults: a systemic review and consensus recommendations for diagnosis and treatment. Gastroenterology 2007; 133:1342–1363.
3. Genevay M, Rubbia-Brandt L, Rougemont AL. Do eosinophil numbers differentiate eosinophilic esophagitis from gastroesophageal reflux disease? Arch Pathol Lab Med 2010; 134:815–825.
4. Fogg MI, Ruchelli E, Spergel JM. Pollen and eosinophilic esophagitis. J Allergy Clin Immunol 2003; 112:796–797.
5. Rothenberg ME. Biology and treatment of eosinophilic esophagitis. Gastroenterology 2009; 137:1238–1249.
6. Whitehorn TB, Liacouras CA. Eosinophilic esophagitis. Curr Opin Paediatr 2007; 19:575–580.
7. Bailey SR, Boustany S, Burgess JK, et al. Airway vascular reactivity and vascularization in chronic airway disease. Pulm Pharmacol Ther 2009; 22:417–425.
8. Chehade M, Sampson HA, Morotti RA, et al. Esophageal subepithelial fibrosis in children with eosinophilic esophagitis. J Pediatr Gastroenterol Nutr 2007; 45:319–328.
9. Szekanecz Z, Koch AE. Mechanisms of disease: angiogenesis in inflammatory diseases. Nat Clin Pract Rheumatol 2007; 3:635–643.
10. Aceves SS, Broide DH. Airway fibrosis and angiogenesis due to eosinophil trafficking in chronic asthma. Curr Mol Med 2008; 8:350–358.
11. Aceves SS, Newbury RO, Dohil R, et al. Esophageal remodeling in pediatric eosinophilic esophagitis. J Allergy Clin Immunol 2007; 119:206–212.
12. Risau W, Flamme I. Vasculogenesis. Ann Rev Cell Dev Biol 1995; 1:73–91.
13. Feltis BN, Wignarajah D, Reid DW, et al. Effects of inhaled fluticasone on angiogenesis and vascular endothelial growth factor in asthma. Thorax 2007; 62:314–319.
14. Fidler IJ, Ellis LM. Chemotherapeutic drugs—more really is not better. Nat Med 2000;6:500–2.
15. Ferrara N. Vascular endothelial growth factor. Eur J Cancer 1996; 32A:2413–2422.
16. Folkman J. Angiogenesis in cancer, vascular, rheumatoid and other disease. Nat Med 1995; 1:27–30.
17. Maniscalco WM, Watkinns RH, D’Angio CT, et al. Hyperoxic injury diseases alveolar epithelial cell expression of vascular endothelial growth factor (VEGF) in neonatal rabbit lung. Am J Respir Cell Mol Biol 1997; 16:557–567.
18. Ferrara N. Vascular endothelial growth factor and the regulation of angiogenesis. Recent Prog Horm Res 2000; 55:15–36.
19. Bhandari V, Choo-Wing R, Chapoval SP, et al. Essential role of NO in VEGF-induced asthma-like angiogenic, inflammatory mucus, and physiologic responses in the lung. Proc Natl Acad Sci USA 2006; 103:11021–11026.
20. Jones N, Iljin K, Domont DJ, et al. Tie receptors: new modulators of angiogenic and lymphangiogenic response. Nat Rev Mol Cell Biol 2001; 2:257–267.
21. Visconti RP, Richardson CD, Sato TN. Orchestration of angiogenesis and atriovenous contribution by angiopoietins and vascular endothelial growth factor. Proc Natl Acad Sci USA 2002; 99:8219–8224.
22. Kishimoto K, Liu S, Tsui T, et al. Endogenous angiogenin in endothelial cells is a general requirement for cell proliferation and angiogenesis. Oncogene 2005; 24:445–456.
23. Gao X, Zu Z. Mechanism of action of angiogenin. Acta Biochim Biophys Sin (Shanghai) 2008; 40:619–624.
24. Koch AE, Polverini PJ, Kunkel SL, et al. Interleukin-8 is a macrophage-derived mediator of angiogenesis. Science 1992; 258:1798–1801.
25. Murdoch C, Monk PN, Finn A. CXC chemokine receptor expression on human endothelial cells. Cytokine 1999; 11:704–712.
26. Salcedo R, Ponce ML, Young HA, et al. Human endothelial cells express CCR2 and respond to MCP-1: direct role for MCP-1 in angiogenesis and tumor progression. Blood 2000; 96:34–40.
27. Nagata M, Yamamoto H, Tabe K, et al. Eosinophil transmigration across VCAM-1-expressing endothelial cells is upregulated by antgen-stimulated mononuclear cells. Int Arch Allergy Immunol 2001; 125:7–11.
28. Furuta GT. Eosinophilic esophagitis: an emerging clinicopathologic entity. Curr Allergy Asthma Rep 2002; 2:67–72.
29. Zanetta L, Marcus SG, Vasile J, et al. Expression of von Willebrand Factor, an endothelial cell marker, is up-regulated by angiogenesis factors: a potential method for objective assessment of tumour angiogenesis. Int J Cancer 2000; 85:281–288.
30. DeLisser HM, Christofidou-Solomidou M, Strieter RM, et al. Involvement of endothelial PECAM-1/CD-31 in angiogenesis. Am J Pathol 1997; 151:671–677.
31. Ferrara N. Role of vascular endothelial growth factor in regulation of physiological angiogenesis. Am J Physiol Cell Physiol 2001; 280:C1358–C1366.
32. Li A, Dubey S, Varney ML, et al. Il-8 directly enhanced endothelial cell survival, proliferation, and matrix metalloproteinases production and regulated angiogenesis. J Immunol 2003; 170:3369–3376.
33. Berry M, Brightling C, Pavrod I, et al. TNF-alpha in asthma. Curr Opin Pharmacol 2007; 7:279–282.
34. Sainson RC, Johnston DA, Chu HC, et al. TNF primes endothelial cells for angiogenic sprouting by inducing a tip cell phenotype. Blood 2008; 111:4997–5007.
35. Cohen MS, Kaufman AB, Palazzo JP, et al. An audit of endoscopic complications in adult eosinophilic esophagitis. Clin Gastroenterol Hepatol 2007; 5:1149–1153.
36. Majno G. Chronic inflammation: links with angiogenesis and wound healing. Am J Pathol 1998; 153:1035–1039.
37. Bagli E, Xagorari A, Papetropoulos A, et al. Angiogenesis in inflammation. Autoimmun 2004; 3(suppl 1):S26.
38. Szekanecz Z, Koch AE. Vascular endothelium and immune responses: implications for inflammation and angiogenesis. Rheum Dis Clin North Am 2004; 30:97–114.
39. Puxeddu I, Ribatti D, Crivellato E, et al. Mast cells and eosinophils: a novel link between inflammation and angiogenesis in allergic disease. J Allergy Clin Immunol 2005; 116:531–536.
40. Hoshino M, Takahashi M, Aoike N. Expression of vascular endothelial growth factor, basic fibroblast growth factor and angiogenin immunoreactivity in asthmatic airways and its relationship to angiogenesis. J Allergy Clin Immunol 2001; 107:295–301.
41. Kristan SS, Molovrh MM, Silar M, et al. Airways angiogenesis in patients with rhinitis and controlled asthma. Clin Exp Allergy 2009; 39:354–360.
42. Kikuchi S, Kikuchi I, Takaku Y, et al. Neutrophilic inflammation and CXC chemokines in patients with refractory asthma. Int Arch Allergy Immunol 2009; 149:87–93.
43. Bates DO. Vascular endothelial growth factors and vascular permeability. Cardiovasc Res 2010; 87:262–271.
44. Koch S, Tugues S, Li X, et al. Signal transduction by vascular endothelial growth factors. Biochem J 2011; 437:169–183.
45. Lukacs NW, Strieter RM, Chensue SW, et al. TNF-alpha mediates recruitment of neutrophils and eosinophils during airways inflammation. J Immunol 1995; 154:5411–5417.
46. Lasalle P, Delneste Y, Gosset P, et al. Potential implications of endothelial cells in bronchial asthma. Int Arch Allergy Appl Immunol 1991; 94:233–238.
47. Gupta SK, Fitzgerald JF, Kondratyuk T, et al. Cytokine expression in normal and inflamed esophageal mucosa: a study into the pathogenesis of allergic eosinophilic esophagitis. J Pediatr Gastroenterol Nutr 2006; 4:22–26.
48. Straumann A, Bauer M, Fischer B, et al. Idiopathic eosinophilic esophagitis is associated with a Th2-type allergic inflammatory response. J Allergy Clin Immunol 2001; 108:954–961.
49. Broide DH, Stachnick G, Castaneda D, et al. Inhibition of eosinophilic inflammation in allergen challenged TNF receptor p55/p75- and TNF-receptor p55-deficient mice. Am J Respir Cell Mol Biol 2001; 24:304–311.
50. Blanchard C, Wang N, Stringer KF, et al. Eotaxin-3 and a uniquely conserved gene expression profile in eosinophilic esophagitis. J Clin Invest 2006; 116:536–547.
51. Bouwmeester T, Bauch A, Ruffner H, et al. A physical and functional map of the human TNF-alpha/NF-kappa B signal transduction pathway. Nat Cell Biol 2004; 6:97–105.
52. Hayden MS, Ghosh S. Shared principles in NF-kappaB signaling. Cell 2008; 132:344–362.
53. Walters EH, Reid D, Soltani A, et al. Angiogenesis: a potentially critical part of remodeling in chronic airway diseases? Pharmacol Ther 2008; 118:128–137.

angiogenesis; angiogenin; eosinophilic esophagitis; esophageal remodeling; IL-8; NFκB; TNF-α; VEGF-A

Supplemental Digital Content

Back to Top | Article Outline
Copyright 2012 by ESPGHAN and NASPGHAN