Identification of a defensin as novel allergen in celery root: Api g 7 as a missing link in the diagnosis of celery allergy?

Abstract

Allergy to celery (Apium graveolens) root (celeriac) is one of the most common food allergies in Northern Europe, with a prevalence of up to 0.45% in the adult population.1 It mostly leads to local oral symptoms, but even severe anaphylactic reactions are reported,2 with a prevalence of up to 6% in Europe among subjects with recorded anaphylactic reactions.3 Six celery allergens are listed by the WHO/IUIS Allergen Nomenclature Sub-Committee4: Api g 1 (PR-10 protein), Api g 2 and Api g 6 (two non-specific lipid transfer proteins), Api g 3 (chlorophyll a/b binding protein), Api g 4 (profilin), and Api g 5 (FAD-containing oxidase). Celeriac allergy is often accompanied by respiratory allergy to mugwort (Artemisia vulgaris) pollen,5 which cannot be explained by IgE cross-reactivity to known homologous celeriac allergens. However, celeriac allergic patients often show sensitization to mugwort defensin Art v 1 (Table S1).6 This prompted us to investigate the potential presence and role of a defensin-related allergen in allergy to celeriac. A celeriac defensin cDNA was cloned (accession# MT380846) and a recombinant (r) Api g defensin was produced. In parallel, mass spectrometric analysis demonstrated the presence of a corresponding defensin-like protein in celeriac extract and confirmed its entire mature amino acid sequence (Figure S1). An alignment with other defensin-like sequences is shown in Figure S2A. IgE-binding capacity and allergenicity of celeriac defensin (officially named Api g 7, www.allergen.org4) was investigated by immunoblotting and ImmunoCAP testing in eight celeriac allergic patients, two celeriac-tolerant controls with mugwort pollen allergy and a non-atopic control. Six of the celeriac allergic patients underwent DBPCFC, showing symptoms ranging from mild subjective reactions to stronger objective symptoms (Table S1), while two patients with a history of anaphylactic reactions to celeriac did not undergo DBPCFC. All celeriac allergic patients were sensitized to rApi g 7 (Figure 1A, lanes 1–8). Levels of specific (s) IgE determined by ImmunoCAP showed no clear correlation between rApi g 7 and celeriac extract, whereas sIgE levels to rApi g 7 and rArt v 1 correlated strongly (Figure 1B, Table S1). Allergenic activity of rApi g 7 was assessed by mediator release (MR) assay, using a humanized rat basophil leukemia cell line. rApi g 7 was able to cross-link FcεRI bound IgE and induced MRmax of around 60% (Figure 1C), a slightly lower level than to rArt v 1 (MRmax = 70%). Api g 7 and Art v 1 share 60% sequence identity (Figure S2A,B) and display common IgE-binding epitopes as demonstrated by IgE cross-inhibition experiments. rArt v 1 was able to completely abolish IgE-binding to rApi g 7 in all of three studied sera (Figure 2A), while rApi g 7 could only partially outcompete IgE-binding to rArt v 1 (Figure 2B), suggesting that Art v 1 may act as primary sensitizer of a cross-reactive response to Api g 7. The observation that five of the eight allergic patients studied here had a negative ImmunoCAP test to celeriac, but were clearly sensitized to rApi g 7 (Table S1), suggested underrepresentation of Api g 7 in that allergen extract. To compensate for a scarcity of Api g 7 in allergen reagents for routine testing, extract spiking or complementary testing with Api g 7 could serve to improve diagnostic sensitivity. In conclusion, our data show that celeriac contains an IgE reactive defensin that likely could explain the previously known association between celeriac allergy and mugwort pollen sensitization. Furthermore, the novel allergen Api g 7 may be useful for enhancing the sensitivity of in vitro diagnostic testing. Further studies in larger patient cohorts will show if an association with more severe reactions to celeriac can be established. The authors thank Elke Haberkorn for performing the Api g 7 ImmunoCAP measurements and Luisa Schwaben for supporting the MS measurements. Open access funding enabled and organized by ProjektDEAL. JL, LM, and HL are employees of Thermo Fisher Scientific. The other authors declare no conflict of interest in relation to this work. 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.

Topics & Concepts