Prostaglandin E Receptor 2 (EP2) Dysregulation in Allergic Fungal Rhinosinusitis Nasal Polyp Epithelium

Abstract

Objectives

Allergic fungal rhinosinusitis (AFRS) is an eosinophilic subtype of chronic rhinosinusitis with nasal polyposis (CRSwNP). This study aimed to investigate the transcriptome of AFRS nasal polyp epithelium.

Methods

Sinonasal epithelial cells were harvested from healthy nasal mucosa and polyp tissue collected from participants undergoing elective sinonasal surgery. Primary epithelial cells were subsequently grown in air/liquid interface and subjected to RNA‐seq analysis, RT‐qPCR, immunoblotting, and immunostaining.

Results

A total of 19 genes were differentially expressed between healthy and AFRS sample epithelium. The second top candidate gene, ranked by adjusted p‐value, was prostaglandin E receptor 2 (PTGER2). The upregulation of PTGER2 was confirmed by RT‐qPCR and immunoblot. The presence of the EP2 receptor, encoded by the PTGER2 gene, was confirmed by immunocytochemistry.

Conclusion

PTGER2 is a potential novel therapeutic target for AFRS. EP2 dysregulation is associated with aspirin‐exacerbated respiratory disease, potentially giving insight into common mechanisms of disease in severe CRSwNP.

Level of Evidence

NA Laryngoscope, 135:S1–S8, 2025

This work represents the first report of dysregulated prostaglandin receptor synthesis in allergic fungal rhinosinusitis (AFRS). In an unbiased analysis of all genes expressed in AFRS versus healthy sinonasal epithelium at air–liquid interface, PTGER2 was the second most significantly upregulated gene. This finding was replicated by quantitative RT‐PCR and immunoblot in patient samples not included in bulk RNA‐seq analysis.

INTRODUCTION

Allergic fungal rhinosinusitis (AFRS) is a recalcitrant subtype of eosinophilic chronic rhinosinusitis with nasal polyposis (CRSwNP) with unmet needs for improved diagnostic and treatment options. ¹ AFRS presents in younger patients relative to other forms of CRSwNP, with an average age <30 years and reports in children as young as 7 years. ² There is an increased prevalence of AFRS among males, with estimates ranging from 1.4 to 2.6: 1. ³ , ⁴ , ⁵ Additionally, African Americans and patients from lower socioeconomic backgrounds with AFRS are more likely to present with signs of advanced CRSwNP, including bony erosion of the neighboring skull base and orbit, as well as facial deformities. ³ , ⁴ , ⁵ Geographically, AFRS is associated with areas of increased ambient temperature and humidity, especially in the Southeastern United States. ⁶

The clinical presentation of AFRS is unique among CRSwNP and is characterized by extensive nasal polyposis in either unilateral or bilateral nares (Fig. 1). ⁷ Among affected sinuses, AFRS is associated with extra‐sinus expansion, marked by the noninvasive encroachment on the orbital and intracranial compartments. This finding is reported in ~30% of cases and represents a common cause for concern by both patients and referring providers alike. ⁸ Interestingly, despite the high burden of CRSwNP associated with AFRS, affected patients can report relatively low sinonasal symptoms. ⁷ The incidence of comorbid asthma is also lower than other forms of CRSwNP, occurring in 23% of patients with AFRS versus >48% in other forms of CRSwNP. ⁹ , ¹⁰ Despite appropriate treatment with intranasal corticosteroids and comprehensive endoscopic sinus surgery, rapid recurrence of nasal polyposis is expected in ~60% of patients. ¹¹ Currently available biologics with indications for CRSwNP have not been adequately studied in AFRS, representing a critical barrier to their use in this population. ¹

Fig. 1.

Clinical characteristics of AFRS. High‐resolution coronal CT with bone window demonstrating characteristic AFRS features. Unilateral disease presentation following prior endoscopic sinus surgery highlights unique presentation with the potential to advance from one sinonasal region to another. Expansion of sinonasal cavities with thickened mucosa (arrows) and dense concretions (asterisks) (A). Eosinophilic, allergic‐type mucin in AFRS. Formation of compact‐layered aggregates of mucin and inflammatory cells and debris are characteristic findings. Fungal elements present in extracellular mucin (arrowheads) (B). Hematoxylin and eosin stain, 200× magnification. Inset: Walls of fungal hyphae appear black when stained with Grocott methenamine silver stain (GMS). GMS stain, 400× magnification. CT, computed tomography.

Strategies to identify AFRS are based on our current understanding of this disease. Unlike other forms of CRSwNP, AFRS is mediated by type‐1 hypersensitivity reactions to noninvasive fungal elements. The most utilized diagnostic algorithm, the Bent and Kuhn criteria, ¹² requires evidence of (1) positive fungal stain and (2) type‐1 hypersensitivity to fungal elements, as well as (3) nasal polyposis, (4) characteristic computed tomography (CT) findings, defined as “serpiginous areas of high attenuation in affected sinuses”, and (5) eosinophilic mucin without fungal invasion as essential components. ⁷ However, these criteria are problematic for several reasons, including the necessity of collecting surgical tissue to confirm disease, as well as the recognition of patients with local sinonasal sensitivity to fungal elements without evidence of systemic reactions. ¹ This presence of local allergy, which is an area of ongoing study, ¹³ , ¹⁴ represents a situation where an affected patient would fail to be identified by current diagnostic methods. In this study, we therefore sought to improve our understanding of the pathognomonic underpinnings of this recalcitrant disease to inform improved strategies for both patient identification and treatment.

MATERIALS AND METHODS

Subject Identification and Tissue Collection

This use of human subjects in this research was approved by the Emory University Institutional Review Board (IRB00102406). Adult patients electing to undergo endoscopic sinus surgery for the management of AFRS, Aspirin‐exacerbated respiratory disease (AERD) or non‐inflammatory skull base pathology (healthy), were recruited from the Sinus, Nasal & Allergy Center at Emory University between 2019 and 2023. Patients were identified as previously described. ¹⁵ AFRS subjects were identified according to Bent and Kuhn criteria. ¹² AERD subjects were identified by a history of CRSwNP, moderate to severe asthma, and a reported history of NSAID intolerance. Informed consent was given at the time of study enrollment. Uninflamed control mucosa was collected from the sphenoid sinus. Tissues were collected in normal saline prior to further processing.

Tissue Processing and Cell Culture

Tissue processing and cell culture were performed as previously described. ¹⁶ Briefly, nasal tissue was minced in a 50‐mL conical tube using dissecting scissors. The small pieces of remaining tissue were placed into digestion media composed of 0.5% w/v protease (Sigma‐Aldrich #P5147) and 10% w/v DNase (Roche, 11284932001) in DMEM/F‐12 (Thermo Fisher Scientific, 11320033). The mixture was placed on a rocker at 4°C for 4 h. The digestion was stopped by adding 10% FBS and tissue filtered through a 100‐μm strainer. The flowthrough was collected, washed in HBSS, and centrifuged at 450 g for 8 min. The pellet was resuspended in red cell lysis buffer and incubated at RT for 5 min. Cells were centrifuged at 450 g for 8 min and the cells were resuspended in conditional reprogramming culture media. The resulting cell suspension was seeded onto cell culture flasks containing irradiated 3T3 fibroblast feeder cells (ATCC, SCRC‐1010) for basal epithelial cell expansion. After expansion, cells were then grown on collagen coated transwells at air/liquid interface for 14 days in E‐ALI media. ¹⁶

Bulk RNA Sequencing

RNA isolation was performed on five healthy and four AFRS individually snap‐frozen cell pellets. RNA extraction was conducted using the miRNEAsy mini kit (Qiagen) and quantified using a Nanodrop Spectrophotometer (Thermo Fisher Scientific). Quality of the RNA was evaluated using RNA 6000 Nano reagents on the 2100 bioanalyzer (Agilent).

RNA sequencing library preparation was performed using the NEBNext Ultra II RNA Library Prep Kit for Illumina according to the manufacturer's recommendations (New England Biolabs). Sequencing libraries were validated on the Agilent 2100 Bioanalyzer System (Agilent Technologies) and quantified using Qubit 2.0 Fluorometer (Invitrogen) as well as by quantitative PCR (Applied Biosystems). The libraries were sequenced on an Illumina sequencer using a 2 × 100 Paired End (PE) configuration to achieve 50 M (100 M total) reads per sample. Raw sequence data (.bcl files) were converted into fastq files and de‐multiplexed using Illumina's bcl2fastq software.

Sequences were quality‐checked using FastQC for completeness, depth, and read quality. Data were trimmed using Trimmomatic to remove adapter contamination. ¹⁷ Sequences were aligned to the HG38 reference genome using STAR aligner. ¹⁸ Gene quantification was done using HTSeq‐count. ¹⁹ DESeq2 was used to determine differentially expressed genes between healthy and AFRS groups. ²⁰ , ²¹ Genes with low counts were filtered by mean normalized counts in DESeq2. Raw p‐values were transformed using the Benjamini–Hochberg correction to account for multiple hypothesis testing. Genes considered significantly differentially expressed were those with adjusted p‐values, also called false discovery rates, <0.05. Graphs were generated using EnhancedVolcano and heatmaps were generated using Graphpad Prism.

RNA Extraction and RT‐qPCR

Total RNA was isolated from cells using an RNeasy mini kit (Qiagen). cDNA was prepared from 1 μg RNA using an iScript cDNA synthesis kit (Bio‐Rad). RT‐qPCR was performed in triplicate using IQ SYBR Green Supermix (Bio‐rad) and quantified using Bio‐Rad CFX Connect Real‐Time System and CFX Manager software (Bio‐Rad). Samples were normalized to RPS18 and expression levels were calculated using the 2^−ΔΔCt method. Real‐time primers were previously characterized and are listed in Table S1. ²²

Immunoblot Analysis

Cell pellets were snap frozen and stored at −80°C until processing. Pellets were resuspended in RIPA lysis buffer (Thermo Fisher Scientific) containing Halt™ Phosphatase and protease inhibitor cocktail (Thermo Fisher Scientific) and incubated on ice for 30 min. Lysates were then centrifuged at 4°C for 10 min at full speed. The supernatant was removed and placed into a clean tube. Total protein was assessed using the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific). For protein electrophoresis, samples were diluted in 4× Laemmli Sample Buffer containing β‐mercaptoethanol and separated on a 7.5% Mini‐PROTEAN® TGX gel (Bio‐rad). Proteins were transferred to a PVDF low fluorescence membrane (Bio‐rad) using the Transblot Turbo (Bio‐rad) at 25 V for 7 min. Membranes were blocked in Everyblot Block (Bio‐rad) for 20 min at RT. Next, membranes were incubated in a 1:500 dilution of rabbit α PTGER2(EP2) (Abcam, ab167171) for 1 h at RT and subsequently washed in TBS+ 0.5% Triton X‐100. Membranes were then incubated in 1:10,000 goat α rabbit IgG StarBright™ Blue 700 (Bio‐rad, 12004162) and 1:5000 α actin hFAB™ Rhodamine (Bio‐rad, 12004163) for 1 h at RT. Membranes were washed in TBS+ 0.5% Triton X‐100 and imaged on the ChemiDoc MP (Bio‐rad). Analysis was performed using Image Lab (Bio‐rad) and densitometric values normalized to actin.

Immunostaining and Imaging

Differentiated sinonasal epithelium on transwell supports were fixed in 4% paraformaldehyde for 15 min at RT, washed three times in Dulbecco's PBS (DPBS), incubated in 0.5% Triton‐x in DPBS for 5 min, followed by 2× 5‐min washes in 5% goat serum in DPBS. Cells were then incubated in primary antibody, or 5% goat serum for the secondary‐only negative control, overnight at 4°C. Next, cells were washed 3 × 5 min in 5% goat serum, incubated in secondary antibody for 1 h at RT, washed 3 × 5 min with 5% goat serum, and washed 3 × 5 min in DPBS. Transwell filters were cut from the insert and mounted on a slide using ProLong Diamond Antifade Mountant w/DAPI (Thermo Fisher Scientific). Antibodies: rabbit α PTGER2(EP2) (Abcam, ab167171), mouse α alpha tubulin (acetyl K40) (Abcam, ab24610), and goat α rabbit Cy2 (Jackson, C840N47). Images were captured using an Olympus IX70 microscope.

Statistical Analysis

Comparisons between groups were performed by Mann–Whitney test or Kruskal–Wallis test using GraphPad Prism software. Data are represented as mean ± standard deviation. Significance is indicated as *p < 0.05 and **p < 0.01.

RESULTS

Participant Demographics and Characteristics

A total of 23 participants were included in this study: nine healthy, nine with AFRS, and five with AERD (Table I). Participants with AFRS were younger than the healthy (29.9 ± 7.5 years vs. 61.9 ± 10.5 healthy, p = 0.000041) and AERD (29.9 ± 7.5 years and 52 ± 6.4 AERD, p = 0.000205) participants. In addition, most participants with AFRS were male (66.7%) and African American (66.7%). Two participants with AFRS also had asthma.

TABLE I.

Subject Demographics and Characteristics.

Cohort	Healthy	AFRS	AERD
n	9	9	5
Age, mean (SD)	61.9 (10.5)	29.9 (7.5)	52 (6.4)
Sex, n (%)
Male	6 (66.7)	6 (66.7)	2 (40)
Race, n (%)
African American	3 (33.3)	6 (66.7)	2 (40)
Asian	2 (22.2)	0 (0)	0 (0)
Caucasian	2 (22.2)	3 (33.3)	3 (60)
Other	2 (22.2)	0 (0)	0 (0)
Ethnicity, n (%)
Hispanic	0 (0)	0 (0)	0 (0)
Asthma, n (%)	0 (0)	2 (22.2)	5 (100)
Prior ESS, mean (SD)	0 (0)	0.4 (0.9)	1.4 (0.9)

ESS = endoscopic sinus surgery; SD = standard deviation.

Identification of Differentially Expressed Genes between AFRS and Healthy Sinonasal Epithelial Cells

To understand the molecular characteristics of AFRS, we first harvested sinonasal epithelial cells from AFRS polyps and healthy sinonasal mucosa, as a control. Figure 2 outlines our method of primary sinonasal epithelial cell isolation, expansion, and differentiation after tissue collection. We extracted RNA and sequenced fully differentiated cell cultures from nine individual participants, providing four AFRS polyp and five healthy sinonasal mucosa samples. Approximately 60 million reads were generated in each sample. We performed differential gene expression analysis on the 22,778 genes with a read count greater than zero. To understand sample variability, we conducted a principal component analysis and found similar gene expression profiles between AFRS samples. The profiles of healthy samples, however, were more heterogeneous and reflective of expected individual variability (Fig. S1). Of the 19 genes we identified as differentially expressed, 12 (0.053%) were upregulated and 7 (0.031%) downregulated (Fig. 3). Overall gene expression in 11 of the differentially expressed genes was low (Fig. 4A). TXN was the only gene highly expressed in all AFRS and healthy samples, and MIR6723 was highly expressed by two healthy samples, but otherwise remained low. FAM43A, PTGER2, and FOLR1 were the genes with the lowest adjusted p‐value; however, PTGER2 expression was the most consistent between samples within each group (Fig. 4A,B), suggesting dysregulation of the gene in AFRS.

Fig. 2.

Cell culture model at air/liquid interface. Illustration depicting culture conditions of primary sinonasal epithelial cells. Cells were grown in flasks containing conditional reprogramming culture (CRC) media and a feeder layer of irradiated 3T3 fibroblasts to promote basal cell expansion (A‐left). Next, the basal cells were transferred to collagen (yellow)‐coated transwell permeable supports and grown at air/liquid interface (ALI) for 14 days to promote differentiation (A‐right). Widefield image of epithelial cell colonies in CRC conditions (B‐left). Dotted circle denotes epithelial cell colony. Image of multiciliated cells (red) present in ALI cultures (B). MCC = multi‐ciliated; MP = mucus producing.

Fig. 3.

Differential gene expression analysis of RNA‐seq data. Volcano plot showing differentially expressed genes by fold change and q‐value (adjusted p‐value). Dotted lines indicate threshold values. The genes selected as differentially expressed are in red. n = 5 healthy and 4 AFRS samples.

Fig. 4.

Expression levels of genes differentially expressed in AFRS. Heatmap representation of genes dysregulated in AFRS. Polyp samples are denoted as P1–P4 and healthy samples as H1–H5. The genes are ranked by adjusted p‐value (A). List of the 19 genes differentially expressed between healthy and AFRS samples (B). n = 5 healthy and 4 AFRS samples.

PTGER2 Expression in Healthy and AFRS Sinonasal Epithelium

The PTGER2 gene encodes for prostaglandin E receptor 2 (EP2), one of four G‐protein‐coupled membrane receptors that mediate the activity of Prostaglandin E₂ (PGE₂). ²³ In addition, dysregulation of PGE₂ receptor expression has been previously characterized in CRSwNP. ²⁴ Specifically, downregulation of EP2 was reported in Aspirin‐exacerbated respiratory disease (AERD), another recalcitrant subtype of CRSwNP. ²⁵ Given the observed twofold increase in PTGER2 expression in AFRS samples compared with controls, along with previous studies suggesting a role for prostaglandin E signaling in nasal polyposis, we investigated the mRNA expression levels of all prostaglandin E receptors in AFRS samples and compared them with those in healthy and AERD samples.

We extracted RNA from differentiated cultures of sinonasal cells from three healthy, four AFRS, and four AERD samples and performed RT‐qPCR to measure PTGER1‐4 expression. We used two new AFRS samples that were not used in our RNA‐seq analysis. We observed ~eight increase in PTGER2 expression in AFRS samples compared with healthy and AERD samples, confirming our RNA‐seq finding (Fig. 5A). There was no difference in PTGER1/4 between healthy, AFRS, and AERD samples (Fig. 5B,C). PTGER3 was undetected in all samples. Next, we investigated EP2 expression by analyzing three healthy and three AFRS samples by immunoblot. These samples were not previously used in any of the RNA expression studies. A twofold increase in EP2 was observed in AFRS samples, compared with healthy controls (Fig. 6A,B). Finally, to visualize EP2 subcellular localization, we performed immunocytochemistry using antibodies generated against EP2 on healthy and AFRS‐differentiated sinonasal epithelial cells. We found perinuclear localization of EP2 in both healthy and AFRS samples (Fig. 6C). This staining is consistent with previous reports in other tissue types such as the epidermis. ²⁶

Fig. 5.

PTGER expression in healthy, AFRS, and AERD sinonasal epithelial cells. RT‐qPCR analysis of PTGER1‐4 mRNA levels in healthy (blue), AFRS (red), and AERD (gray) sinonasal epithelial cells grown at ALI (A–C). n = 3–4 participant samples per cohort. Data are represented as mean ± standard deviation. Kruskal–Wallis test. **p < 0.01.

Fig. 6.

EP2 expression and protein localization in healthy and AFRS sinonasal epithelium. Immunoblots showing EP2 and actin (loading control) in healthy and AFRS sinonasal epithelial cells grown at ALI (A). EP2 levels were significantly increased in AFRS compared with healthy control samples (B). EP2 values were normalized to actin. n = 4 participant samples. Data are represented as mean ± standard deviation. Mann–Whitney test. *p < 0.05. Immunostained healthy (top) and AFRS (middle) sinonasal epithelial cells shows localization of EP2 (green) and nuclei (blue) (C). The secondary‐only negative control shows background staining. Arrow indicates perinuclear staining. Box indicates area magnified. Scale bar is 20 μm. n = 3 participant samples.

DISCUSSION

This work represents the first report of dysregulated prostaglandin receptor synthesis in AFRS. In an unbiased analysis of all genes expressed in AFRS versus healthy sinonasal epithelium at air–liquid interface, PTGER2 was the second most significantly upregulated gene. This finding was replicated by quantitative RT‐PCR in independent patient samples. Expression of the associated EP2 receptor was also increased twofold in participant samples not included in bulk RNA‐seq analysis.

Although this work presents novel insight into the inflammatory background associated with AFRS, it is not the first report of eicosanoid dysregulation in eosinophilic CRSwNP. AERD, a comparably recalcitrant form of CRSwNP, is defined by a pathognomonic under‐expression of EP2 as well as its associated prostaglandin E (PGE) agonist and cyclooxygenase‐2. ²⁷ Mouse models of AERD are created by knocking out the microsomal isoform of PGE₂ synthase‐1 (ptges ^−/− ), thereby depleting PGE₂ and replicating many clinical features of this disease. ²⁸ Unlike AFRS, which is understood to result from IgE‐mediated hypersensitivity reactions, the clinical symptoms of AERD are mediated by this eicosanoid dysregulation. Upon ingestion of cyclooxygenase‐1 inhibiting nonsteroidal anti‐inflammatory drugs, a clinically significant shunting of arachidonic acid from PGE production to pro‐inflammatory cysteinyl leukotrienes occurs. This presents clinically with diagnostic anaphylactoid (i.e., non‐IgE mediated) respiratory reactions, acute gastrointestinal distress, and/or dermatologic rashes. It is interesting to discover the shared association of eicosanoid dysregulation in both AFRS and AERD given the recent report of two patients meeting diagnostic criteria for both forms of CRSwNP. ²⁹

This work highlights the importance of maintaining an appropriate balance of PGE₂ and its associated EP2 receptor in the upper airways. Dysregulation in either direction is now associated with recalcitrant forms of CRSwNP. Although low expression of PGE₂/EP2 is associated with AERD, we now have evidence of EP2 over‐expression in AFRS. This finding has the potential to be clinically relevant as we continue efforts to develop disease‐specific diagnostics and targeted interventions to better control chronic airway disease. Importantly, this work may also provide evidence supporting the role of respiratory biologics such as dupilumab for the treatment of AFRS. A nested analysis of participants with varied CRSwNP subtypes enrolled in a phase II clinical trial of dupilumab versus placebo for the treatment of CRSwNP revealed the greatest improvements in both patient‐reported and objective disease‐specific outcomes among those with AERD. ³⁰ Additionally, it has recently been hypothesized that the mechanism for this increased effect in AERD is subsequent to normalization of PGE₂/EP2 production. ³¹ Of note, patients with AFRS were excluded from this parent trial as well as subsequent pivotal phase III studies. Additionally, the increased expression of EP2 in AFRS may explain the relatively low prevalence of comorbid asthma in this population. Although PGE₂ signaling is generally considered pro‐inflammatory, its function in the lower airway is protective. ³²

The analysis of EP2 expression in AERD samples versus control sinonasal mucosa in this study failed to detect downregulation. One hypothesis deserving additional study is that EP2 is differently expressed in various sinonasal tissues. In a 2006 study, Ying et al measured EP2 expression in the epithelium of nasal biopsy from patients with aspirin‐sensitive rhinosinusitis, compared with healthy controls, and found EP2 upregulated in ciliated epithelial cells and no change in goblet cells. ³³ By contrast, prior work identifying EP2 downregulation in AERD is limited to cultured fibroblasts and whole tissue samples. ²⁵ , ³⁴ , ³⁵ Future study characterizing EP2 expression throughout the upper and lower airways in health and disease is needed.

There are several limitations to our study. The first is our small sample size. We used samples collected from subjects undergoing elective endoscopic sinus surgery at the Sinus, Nasal & Allergy Center at Emory University, limiting the generalizability of our findings. Our studies were conducted using primary cells that were cultured at air/liquid interface to recapitulate the native sinonasal epithelium. As a result, our findings are restricted to the nasal polyp epithelium, which represents only a portion of the intact tissue. Furthermore, cultured cells are not subjected to the native tissue microenvironment. Finally, additional study is needed to evaluate the functional activity of EP2 in CRSwNP.

CONCLUSION

Our results characterize PTGER2 (EP2), a mediator of prostaglandin signaling, dysregulation in cultured AFRS polyp epithelial cells grown at ALI. Our findings identify EP2 as a potential biomarker of AFRS and contribute to the growing knowledge of aberrant eicosanoid signaling in nasal polyposis.

Supporting information

Table S1. Primers used for RT‐qPCR PTGER1‐4 primers were previously characterized.²²

Figure S1. Principal component analysis of RNA‐seq data.