Whole Blood Transcriptomics Identifies Differences in Innate Immune Pathway Expression in Infants at Risk for Peanut Allergy

Peanut allergy affects up to 2%–4% of children in the United States [1]. Despite advances in peanut allergy prevention through early introduction [2], there is limited understanding of molecular processes that drive development of peanut allergy. Similarly, differences between infants who tolerate peanut but demonstrate allergic sensitization and those with clinical reactivity to peanut are poorly understood. In a pilot study by our group, infants at high risk for peanut allergy demonstrated differential gene expression by clinical phenotype, but the magnitude of differences was not clinically significant given the variance of expression [3]. Further exploration of gene networks from differentially expressed gene (DEG) signatures may provide better targets for future studies. The primary objective of this study was to analyse the transcriptomic signature of a cohort of infants at risk for peanut allergy utilising whole blood RNA sequencing. We hypothesized that infants with peanut allergy would have differential expression of genes involved in immune pathways when compared to infants tolerant of peanut, with or without sensitization to peanut.

Infants aged 4–11 months (n = 70) with peanut allergy risk factors (egg allergy, moderate to severe atopic dermatitis, or both) were recruited from Ann and Robert H. Lurie Children's Hospital of Chicago between 2018 and 2021. Subjects were classified as either peanut non-allergic (PNA), peanut sensitised (PS), or peanut allergic (PA). PA subjects had a convincing clinical history of reaction to peanut and peanut skin prick (SPT) wheal ≥ 3 mm or peanut SPT wheal ≥ 8 mm without history of consumption. Subjects regularly consuming peanut without history of an adverse reaction to peanut or who had never consumed peanut and had peanut SPT wheal size of 0–2 mm were classified as PNA. Infants who had ingested peanut without clinical reaction or passed a peanut oral food challenge (OFC) and had a peanut SPT wheal size of 3–7 mm or positive peanut specific IgE (sIgE) were classified as PS.

RNA-seq was performed on whole blood. Deconvolution of gene expression by immune cell type showed no clinically significant differences between peanut allergy groups. Comparison of gene expression profiles by peanut allergy status was carried out using DESeq2 [4]. DEGs were split into upregulated and downregulated groups and were analysed separately in pathway analyses performed using Metascape [5]. Significant differences in single gene expression between groups were reported if there was ≥ ± absolute log₂ fold change of 0.3 and FDR-corrected p ≤ 0.05.

Half of the population was PNA (n = 35, 50%), 12 subjects (17%) were PS, and 23 subjects were PA (33%). Most subjects in each sub-group were male and had eczema. When adjusted for multiple comparisons, there were no statistically significant DEGs between PA and PNA subjects and PA and PS subjects. CLEC12B, a C-type lectin receptor exclusively expressed on skin mast cells, was up-regulated in PS versus PNA subjects (fold change 1.3, FDR p-value 0.012). Pathway analyses demonstrated upregulation of innate immune response, neutrophil degranulation, and cytokine signalling in PS as compared to PNA subjects. Inflammatory responses, innate immune responses, neutrophil degranulation, and regulation of immune effector processes were all down-regulated pathways in PA versus PS subjects (Figure 1).

While there were few singly DEGs between peanut allergy groups, pathway analyses revealed several uniquely downregulated pathways in subjects with peanut allergy. The most highly down-regulated pathways in infants with peanut allergy as compared to those sensitised but tolerant to peanut were inflammatory and innate immune responses. The role of the innate immune response in molecular recognition of food allergens and the development of food allergy has been previously described [6, 7]. Our results of decreased innate immune signatures in PA subjects are in line with findings from studies of early life development of wheeze and asthma. Consensus from childhood allergic asthma studies has demonstrated shifts to Th2 immunity and decreased innate immune gene expression, likely because of early imbalance in Th2/Th1 immunity or a general decrease in host defence capability, which may underlie susceptibility to allergic disease [8]. These findings highlight the necessity to understand the timing and nature of immune developmental defects associated with impaired innate immune response and development of food allergy. Further dissection in a longitudinal study to determine which genes are drivers for our pathway results will be important.

This study was not without limitations. Our classification of peanut allergy was primarily based on reported clinical history, and not all patients underwent OFC to confirm peanut allergy status. We did not have samples to perform single cell RNA sequencing and may have missed some unique DEGs in single cell populations. However, previous studies have demonstrated the utility of using whole blood for RNA extraction, finding associations with genes which would not have been expressed in single cell data [3, 9]. Limitations on blood volume collection in infants restricted additional sample collection for confirmatory functional testing.

Elucidating biologic pathways that distinguish infants who are sensitised but clinically tolerant to peanut from those infants who are truly peanut-allergic is necessary to help implement allergy prevention and treatment strategies. This study is one of the largest to examine whole blood transcriptomic expression profiles in infants at high risk of peanut allergy. These results suggest that disruption of the innate immune response may play a role in peanut allergy development. It is increasingly recognised that control and coordination of innate immunity is needed to maintain equilibrium with appropriate responses to viral infections, immune tolerance, and prevention of atopy. Further investigation of the innate immune response in infants at high-risk of peanut allergy may have implications for the development of clinical reactivity in sensitised individuals.

Abigail Lang assisted with interpretation of the data and drafted the manuscript. Lauren Gunderman and Elizabeth Lippner assisted with critical appraisal of the analyses and revised the manuscript. Samantha Gadd and Matthew J. Schipma analysed the RNA-sequencing data and performed bioinformatics. Ashley Devonshire assisted with initial study design and acquisition of the original data and interpretation of findings, in addition to providing critical appraisal of the analyses and manuscript. Sergejs Berdnikovs provided critical review and interpretation of the results and contributed to writing the manuscript. Rajesh Kumar supervised the analysis and interpretation of results with critical review of the intellectual content of the manuscript. All listed authors have given final approval of the submitted version of the manuscript and agree to be accountable for all aspects of the work related to its accuracy and integrity.

The authors declare no conflicts of interest.