Single‐Cell Profiling Identifies <scp>TNF</scp> ‐Responsive Circulating Eosinophils in Allergic Rhinitis

Allergic rhinitis (AR) is a chronic inflammatory disease driven by type 2 immune responses. Eosinophils, key effector cells involved in AR pathogenesis, are closely associated with symptom severity [1]; however, the functional heterogeneity and distinct activation states of circulating eosinophils in AR remain poorly defined. In this study, we aimed to resolve this heterogeneity at single-cell resolution to identify distinct subsets and their potential roles in AR pathogenesis using an optimized single-cell RNA sequencing (scRNA-seq) protocol for fixed peripheral blood leukocytes from patients with AR and healthy controls, complemented by complete blood count (CBC) analyses and immunofluorescence (IF) staining of nasal mucosal tissues. The full methodological details are available in the Supporting Information. This study included eight patients with AR and seven healthy controls. Compared with controls, patients with AR had significantly higher blood eosinophil counts (p = 0.008), blood eosinophil percentages (p = 0.001), and serum total IgE levels (p = 0.002; Figure 1A, Table S1). scRNA-seq analysis (Figure 1B,C) further revealed a significantly greater proportion of circulating eosinophils in the patients with AR than in the controls (p = 0.029; Figure 1D). These findings were consistent with the CBC results in an independent AR cohort (p < 0.001; Figure 1E) and a previously published dataset [2] (p = 0.002; Figure 1F). Together, these findings provide evidence for a substantial increase in circulating eosinophils in AR patients, underscoring their potential involvement in the disease pathophysiology. As a secondary finding, we also observed an increased proportion of classical monocytes in AR compared with controls (p = 0.021; Figure 1D), which is consistent with a previous report and suggests a broader inflammatory monocyte involvement in AR [3]. To investigate the functional heterogeneity of these expanded eosinophils, we analyzed the 946 eosinophils and revealed two distinct subpopulations (Eos_C1 and Eos_C2; Figure 2A), both of which expressed canonical eosinophil markers (CLC, CCR3, and IL5RA; Figure 2B). The Eos_C2 subset was significantly enriched in patients with AR compared with controls (p = 0.040; Figure 2C,D). Cell cycle analysis revealed a lower proportion of Eos_C2 cells than Eos_C1 cells in S phase (p = 0.026; Figure 2E,F), suggesting that Eos_C2 cells represent a more differentiated, non-proliferative phenotype. We performed differential gene expression analysis to molecularly characterize these two subsets, revealing marked transcriptional divergence between them (Figure 2G). Eos_C1-upregulated genes were associated with proliferation and migration, whereas Eos_C2-upregulated genes were enriched in immune responses, particularly tumor necrosis factor (TNF) responsiveness (Figure 2H). Protein–protein interaction analysis revealed TNF as a central hub among Eos_C2-upregulated genes (Figure 2I). Furthermore, gene set enrichment analysis confirmed significant enrichment of the TNF-mediated signaling pathway in eosinophils from patients with AR versus controls (normalized enrichment score = 1.61; p = 0.005; Figure 2J). To validate the robustness of this TNF-responsive signature, we analyzed two independent public scRNA-seq datasets [4, 5] and observed a similar Eos_C2-like transcriptional program in both (Figure S1). Finally, for further validation at the histological level, we performed IF staining of nasal mucosal sections. The results revealed significantly greater co-localization of the TNF pathway components TRADD and TRAF5 with the eosinophil marker EPX in AR tissues compared to controls (p = 0.0002 and p = 0.0002, respectively; n = 10 per group, five fields per sample; Figure 2K,L), providing direct support for active TNF signaling in tissue-infiltrating eosinophils in AR. Representative negative and isotype control staining performed on both AR and control tissues showed no background fluorescence, confirming the specificity of the immunofluorescence results (Figure S2). To our knowledge, this is the first study to characterize functional heterogeneity among circulating eosinophils in patients with AR at single-cell resolution. We identified a distinct AR-associated eosinophil subset (Eos_C2) with pronounced TNF-responsive activity, low proliferative activity, and a transcriptional profile suggestive of a tissue-ready phenotype. The marker profile of the Eos_C2 subset provides important clues to its potential function. Eos_C2 exhibited elevated expression of ITGAX and TMSB4X, markers linked to migration and cytoskeletal remodeling and have previously been associated with tissue eosinophils in gastrointestinal [6] and airway [7] inflammation, suggesting a potential for recruitment to tissues. Eos_C2 also showed upregulated expression of pro-inflammatory and chemotactic genes, including ALOX5, IL3RA, S100A9, IL16, and CKLF. These features suggest that Eos_C2 may represent a circulating primed subset with both tissue-recruitment potential and pro-inflammatory activity. The Eos_C2 transcriptional program is enriched for TNF-responsive genes, although several highlighted genes are also shared with other inflammatory pathways. Consistent with our findings, TNF-α-deficient mice exhibit reduced eosinophil recruitment in response to AR [8], underscoring the importance of TNF signaling in disease pathogenesis. While current clinical guidelines do not include TNF-pathway–related inhibitors in the management of AR, our data suggest that this pathway may represent a potential therapeutic target. Previously tested TNF inhibitors or agents targeting downstream effectors could therefore be further explored to modulate pathogenic eosinophil activation, especially in patients with eosinophil-driven, refractory disease. Interleukin(IL)-5 is the central survival factor for eosinophils, and both subsets in our data uniformly expressed IL5RA. Eos_C2 may represent a TNF-responsive eosinophil subset whose activation program is superimposed on an IL-5–maintained survival context. Clinically, anti-IL-5/IL-5Rα therapy would be expected to deplete Eos_C2. Prior work has shown that TNF can prolong eosinophil survival independently of IL-5 [9], while IL-5 and TNF may also interact at the transcriptional level [10]. These findings highlight the complex and context-dependent interplay between IL-5 and TNF in shaping eosinophil biology, which warrants further experimental investigation in Eos_C2. Although the overall cell-type resolution and relative expression patterns in our dataset were well preserved, fixation may reduce transcript recovery and introduce bias against low-abundance transcripts. Future studies using independent experimental methods will be important to further confirm the functional relevance of the identified subpopulations. In summary, we identified a TNF-responsive circulating eosinophil subset in patients with AR. These findings provide new insights into eosinophil heterogeneity in AR and suggest potential therapeutic opportunities targeting TNF signaling in eosinophil-driven airway inflammation. S.C. and Y.Z. performed data analysis and prepared the draft of the manuscript and figures. S.Y. conducted the experimental work and contributed to data acquisition and analysis. P.Z. supervized the analytical approaches, provided methodological guidance, and reviewed the data interpretation. H.L. and L.Z. conceived and designed the study, managed and coordinated the research planning, and critically revised the manuscript. All authors approved the final version for submission. The authors declare no conflicts of interest. All data supporting the findings of this study are available in the main text or the Supporting Information. The processed scRNA-seq data have been deposited in the OMIX repository of the China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences at https://ngdc.cncb.ac.cn/omix/, reference number OMIX012038. FIGURE S1: Independent validation in two public scRNA-seq datasets. FIGURE S2: Negative and isotype controls for immunofluorescence staining. TABLE S1: Clinical characteristics of AR and control groups. Data were presented as median (Q1, Q3) or n (%). AR, allergic rhinitis; IgE, immunoglobulin E; NA, not available. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.