C70-11 T1-T2 Interaction and Aberrant Keratinization in the Airways Distinguishes Severe From Mild-Moderate Asthma

Abstract Rationale Asthma is a heterogenous disease afflicting over 300 million people worldwide. Patients with mixed type-1 (T1) /type-2 (T2) immune response in the airways are recognized to be the sickest. The aim of this study was to determine the differential roles of Type 1 and Type 2 immune responses observed in a cohort of treatment-refractory severe asthma patients. Methods Wild type (WT) and Il4ra-/- mice (Il4ra-/- KO) were subjected to either mixed T1/T2 severe asthma (SA) or T2 dominant mild-moderate (MMA) asthma model. Airway hyperresponsiveness (AHR) to increasing methacholine dose was assessed using Flexivent technology. Mouse lungs were either used for histology (H&E, PAS or Masson’s Trichrome staining), or quantitative RT-PCR or digested for single cell preparation for Flow cytometry. SLIDE, a novel interpretable machine learning algorithm, was used to assess the airway epithelium transcriptome from participants in the Immune Mechanisms in Severe Asthma (IMSA) cohort. Results SAWT and SAIl4ra-/- KO mice and MMAWT mice showed elevated inflammation in the lung as compared to lack of inflammation in the MMAIl4ra-/- KO mice, showing the dominant role of T2 cytokines in the inflammatory response in the MMA model. Mucus production was completely ablated in Il4ra-/- KO mice in both models. There was no difference in collagen deposition in the lungs of SAWT and MMAWT mice as revealed by Masson’s Trichrome staining. AHR measurements showed differential outcomes in the SA vs MMA model in the absence of IL-4Rα signaling compared to WT mice, with the KO mice showing significant reduction in both RN (Newtonian resistance) and Rrs (Respiratory system resistance) in the MMA model, but with only Rrs significantly reduced in the SA model with no change in RN. Analysis of the airway transcriptome by SLIDE unraveled significant differences in gene expression programs between SA and MMA patients with 5 significant latent factors (LFs) of disease corresponding to context-specific gene co-expression programs. Subsequent pathway analysis of these significant LFs by gene set enrichment analysis revealed enrichment of keratinization in SA when compared to HC or MMA as well as increases in O-glycosylation of mucins and arachidonic acid metabolism when compared to MMA. Conclusions While T2 inflammation regulates mucus production in both the T2 dominant MMA and mixed T1high/T2SA model, it differentially regulates lung inflammation and AHR in these models. Increased airway remodeling in SA compared to MMA as revealed by SLIDE is not due to differential extracellular collagen deposition but increased keratinization as suggested by SLIDE. This abstract is funded by: This study was supported by the National Institutes of Health grants P01 AI106684 (to A. R. and S. E. W. ), R35 HL166219 and R01 AI176632 (to A. R. ), K08 HL164887 (to M. C. G. ), R01 HL153058 (to S. E. W. ) and DP2AI164325 (to J. D. ). This research was supported in part by the University of Pittsburgh Center for Research Computing and Data, RRID: SCR₀22735, through the resources provided. Specifically, this work used the HTC cluster, which is supported by NIH award number S10OD028483.