In vitro vaccine challenge of PBMCs from BNT162b2 anaphylaxis patients reveals HSP90α‐NOD2‐NLRP3 nexus

mRNA vaccines are highly effective at eliciting protective immunity against SARS-CoV-2. However, rare serious adverse events (SAEs), such as anaphylaxis, are estimated to occur at 5 cases/million doses, which is 2x-5x that of traditional vaccines.1 As mRNA vaccines are novel, the triggers and mechanisms of vaccine-associated anaphylaxis are not known. Anaphylaxis is classically elicited by antigen-specific IgE activation of mast cells and basophils, while IgG-driven alternative pathways by direct activation of mast cells and complement activation are also known. In BNT162b2 vaccine, the mRNA-coding SARS-CoV-2 spike protein is enclosed in a lipid nanoparticle comprising cholesterol, 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC), ALC-0315 ([4-hydroxybutyl] azanediyl) bis(hexane-6,1-diyl) bis(2-hexyldecanoate), and ALC-0159 (2-[[polyethylene glycol]-2000]-N,N-ditetradecyl acetamide). PEG lipid is reported to be a possible antigenic source, which either interacts with IgE directly to precipitate in the classical anaphylaxis pathway or indirectly through activation of the complement system. In this study, adjudicated cases of anaphylaxis (Brighton collaboration level of 1 or 2, n = 7) and individuals with minimal reaction (n = 8) (Table S1) were compared for relative expression of allergy and inflammation-related cytokines, and the presence of anti-PEG2000 IgE and IgG in plasma. The study was approved by the local ethics committee (DSRB 2012/00917) with written consent provided by participants. All anaphylaxis patients developed symptoms with their first dose of BNT162b2, and none had prior anaphylaxis, though 6/7 had an atopic history of allergies. elisa readouts of plasma total IgE (Figure S1A), antigen-specific IgE (Figure S1B), and IgG (Figure S1C) against PEG2000 or the SARS-CoV2 spike protein were, however, not elevated in anaphylaxis cases. As both IgE and IgG to PEG showed no significant difference between cases and controls, this prompted us to investigate other signaling pathways that may trigger anaphylaxis. PBMCs were seeded in a 24-well plate at 2 × 106 cells/well in 1 ml of RPMI-1640, 20% autologous plasma, and 1 μg/ml of BNT162b2. Cells were harvested at 0, 2, and 6 h and total RNA extracted and assayed using RT2 Profiler PCR Arrays (Qiagen). A distinct set of innate immune response genes were significantly induced in the anaphylaxis group at 2 h poststimulation, which was amplified at 6 h poststimulation (Figure 1A,B). Of note, genes (n = 66) downstream of Ig-dependent mechanisms of allergy were not detectable (Table S2). Pathway analysis for genes specifically upregulated in cases by ≥40% (Log2 fold-change ≥0.5) with p-value <0.05 (Figure 1C) showed that they form an interconnected network linking NOD-like receptor signaling2 (FDR = 8.56e−15), NFκB signaling (FDR = 1.36e−09), response to LPS (lipopolysaccharide) (FDR = 2.08e−08), and inflammatory response (4.05e−08). The network suggests that the most significantly induced gene, HSP90AA1 (2 h FC = 1.43, p < 0.001; 6 h FC = 1.69, p < 0.001), which codes for the stress-induced molecular chaperone HSP90α,3 is integral to these pathways that lead directly to NLRP3 inflammasome activation.4 This is in agreement with NLRP3 inflammasome activation by nanoparticles and cationic lipids.5 Confirmatory qPCR-verified NLRP3, NOD2, and HSP90AA1 (Figure 2A,B) were specifically upregulated in the anaphylaxis group and could exacerbate inflammation through active transcription of multiple proinflammatory mediators (Figure 2B) and NLRP3 inflammasome. The pathway map (Figure S2A,B) suggests a key role of HSP90α in activating NOD2 and TLR4 signaling, as well as the NLRP3 inflammasome for the anaphylactic reaction, as the summation of their fold-changes are very well-correlated (R2 = 0.97, p = 0.008) (Figure S2C). By contrast, other regulators of NLRP3 inflammasome activity (Figure S3A) and Ig-dependent pathways were not differentially expressed (Figure S3B). Our findings here showed a lack of correlation between anti-PEG IgE and anaphylaxis, although lipid-conjugated PEG needs further investigation.6 Instead, we found a cellular basis underpinning the anaphylaxis reaction due to the activation of a potent inflammation pathway. It is a starting point for future work on the possible role of cells such as Tregs and specific cytokines. We would like to thank Yong Jie Tan and Zhiyi Goh for their technical assistance, and the HSA Expert Panel on Hypersensitivity Reactions for the adjudication of cases. This work was supported by funding from SIgN core fund F0006; National Medical Research Council (NMRC) COVID-19 research fund (COVID-19RF-11, COVID-19RF-18, and COVID-19RF-60). The authors declare no conflict of interest. Informed consent (DSRB 2012/00917) was obtained for all subjects involved in this study. Appendix S1 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.