Identification of Ulocladium chartarum as an important indoor allergen source

To the Editor, The constant exposure to fungal spores, which constitute the largest proportion of aerobiological particles, can cause severe health problems, including allergic diseases.1,2 Appropriate management of fungal allergies is hampered by the fact that our knowledge about fungal allergy is still limited to a small number of thoroughly investigated fungi, whereas for the majority of species, it is not yet known whether they play a role in allergic diseases.2 This lack of knowledge together with unreliable diagnostic results obtained with fungal allergen extracts of poor quality contributes to a general underdiagnosis of fungal allergy.3,4 In the present study, we investigated the allergenic potential of Ulocladium chartarum, an opportunistic human pathogen5 that can grow on various substrates.6,7 Exposure to Ulocladium species often occurs in the indoor environment as they are commonly detected inside damp buildings and are regarded as an indicator of water damages.6– 9 The fact that U. chartarum belongs


Identification of Ulocladium chartarum as an important indoor allergen source
To the Editor, The constant exposure to fungal spores, which constitute the largest proportion of aerobiological particles, can cause severe health problems, including allergic diseases. 1,2 Appropriate management of fungal allergies is hampered by the fact that our knowledge about fungal allergy is still limited to a small number of thoroughly investigated fungi, whereas for the majority of species, it is not yet known whether they play a role in allergic diseases. 2 This lack of knowledge together with unreliable diagnostic results obtained with fungal allergen extracts of poor quality contributes to a general underdiagnosis of fungal allergy. 3,4 In the present study, we investigated the allergenic potential of Ulocladium chartarum, an opportunistic human pathogen 5 that can grow on various substrates. 6,7 Exposure to Ulocladium species often occurs in the indoor environment as they are commonly detected inside damp buildings and are regarded as an indicator of water damages. [6][7][8][9] The fact that U. chartarum belongs | 3203 LETTERS to the same family as the outdoor aeroallergen source Alternaria alternata suggests that U. chartarum might also represent an allergen source. 6,7 IgE immunoblots, performed with sera from patients sensitized to different mold species (Table S1), demonstrated the high allergenic potential of U. chartarum, since patients' IgE antibodies recognized several U. chartarum proteins ( Figure 1A), whereas exposure of the blotted proteins to sera from non-allergic individuals or to immunodetection reagent did not lead to any unspecific binding (data not shown). Interestingly, despite their phylogenetic relationship, the IgE-binding profile of A. alternata ( Figure S1) differed significantly from the one of U. chartarum. Mold-allergic patients not only recognized more proteins in U. chartarum than in A. alternata extract, but also showed an overall stronger reactivity to them.
Peptide mass fingerprinting allowed to identify a 15 kDa U. chartarum protein (A in Figure 1A In case of rUlo c 1, mainly beta-sheet structures (minimum of far-UV spectrum at 213 nm) were found, whereas predominantly alphahelical structures (minima at 207 nm and 220 nm) were detected for rUlo c 2 and rUlo c 6 ( Figure S2).
ELISAs performed with sera from mold-allergic patients showed the IgE-binding capacity of the three recombinant allergens, with rUlo c 1 displaying the highest levels of IgE reactivity ( Figure 1B). Furthermore, IgE inhibition immunoblots, where nitrocelluloseblotted protein extracts from U. chartarum were exposed to serum pools that had been pre-incubated with the recombinant proteins, showed that the recombinant allergens were able to completely inhibit patients' IgE binding to their natural counterparts ( Figure 1C). This indicates that the recombinant allergens represent well-folded proteins which contain all the IgE-binding epitopes present in their natural counterparts and suggests that rUlo c 1, rUlo c 2, and rUlo c 6 could be used as tools for in vitro diagnosis of U. chartarum sensitization.
The prevalence of the molecules' IgE reactivity was analyzed by ELISA using sera from 85 individuals sensitized to different mold species ( Figure S3). rUlo c 1 was recognized by 58% of the patients, indicating that this molecule represents a major mold allergen, whereas rUlo c 2 was recognized by 43% and rUlo c 6 by 40% of the patients, suggesting minor mold allergens.
Sequence comparison of the three U. chartarum allergens with homologous A. alternata proteins revealed protein sequence identities of 89% between Ulo c 1 and Alt a 1, 99% between Ulo c 2 and its homologous A. alternata protein, and 100% between Ulo c 6 and Alt a 6. The high sequence homology between Ulo c 1 and Alt a 1 and the suggested presence of Alt a 1-homologous proteins in other species of the Pleosporaceae family 10 prompted us to investigate the potential cross-reactivity between rUlo c 1 and rAlt a 1 in IgE inhibition immunoblots. As depicted in Figure 2A A. alternata showed that the allergens were all immediately released from the spores ( Figure S4). These findings suggest that rapid allergen elution might also occur when spores reach mucosal surfaces, where they can then elicit allergic reactions.
In conclusion, in this study we provide evidence that U. chartarum represents an important, so far underestimated, allergen source that shows at least partial cross-reactivity to the evolutionarily related species A. alternata. Owing to the fact that in contrast to the outdoor mold A. alternata, U. chartarum is an important component of the indoor environment, it is especially important to raise the awareness that U. chartarum represents a potential cause of respiratory allergic diseases and to include the species into routine allergy diagnosis. We identified the species' first allergens, Ulo c 1, Ulo c 2, and Ulo c 6, and produced them as IgE-reactive recombinant molecules. These recombinant allergens will expand the repertoire of fungal allergens available for fungal allergy diagnosis, thus improving diagnosis' specificity and sensitivity.
The description of the used methods can be found in the supporting information.

F I G U R E 1 Identification and characterization of IgE-reactive proteins from Ulocladium chartarum. (A) Nitrocellulose-blotted proteins from
U. chartarum were exposed to two pools of sera [pool 1: sera 1, 2, 45-47, and 74; pool 2: sera 3-4, 13-14, 28-31, 34-36, and 58-61] from mold-allergic patients. IgE-reactive proteins identified by mass spectrometry are marked with A, B, and C; strongly stained but unidentified bands are marked with D, E, and F. Molecular weight markers are indicated in the left margins. (B) Determination of the IgE-binding capacity of the recombinant allergens rUlo c 1, rUlo c 2, and rUlo c 6 by ELISA using sera from mold-allergic patients. Results are displayed as mean OD values, and standard deviations from two technical replicates are indicated as error bars in the bar charts. The red line represents the cutoff value, which was calculated from the mean OD values plus two times the standard deviation of sera from three non-allergic patients. Sera that gave OD values above 3.0 are marked with an asterisk (*). (C) Inhibition immunoblot: Blotted proteins of U. chartarum protein extracts were exposed to pools of sera (rUlo c 1: sera 25-27, 42, and 78; rUlo c 2: sera 15, 28-32, and 57; rUlo c 6: sera 3-4, 13-14, 28-31, 34-36, and 58-61) that had been pre-incubated with rUlo c 1, rUlo c 2, or rUlo c 6 or, for control purposes, with buffer only. Molecular weight markers are indicated in the left margins. Arrows point to reduction in IgE binding to natural Ulo c 1, Ulo c 2, and Ulo c 6 after preincubation with the recombinant proteins Dogs (Canis lupus familiaris) spontaneously develop AD which is strikingly similar to its human counterpart and can be used to answer questions that may be challenging when studying human patients. 5,6 Canine IL-31RA mRNA splice variants encode only three isoforms, two full-length and one truncated in the extracellular domain, which correspond to human homologs identified in Cevikbas' upregulated subset. As in the human full-length IL-31RA, the extracellular region of the canine long IL-31RA isoforms includes a signal peptide and five fibronectin type III (FnIII) domains, the first two of which include the cytokine-binding domain in human IL-31RA. 1,2 The short canine IL-31RA is missing the signal peptide and the first FnIII domain, as in its non-functional human homolog. 1 In functional studies of the human homologs of these IL-31RA splice variants by Maier and colleagues, translation of the short isoform was shown to be markedly reduced, and when expressed it was not able to activate STAT signaling in response to IL-31, suggesting a negative influence on IL-31 signaling. 1 The potential role of these transcript variants during allergic inflammation can therefore be clarified by examining their transcription in atopic dogs.
The atopic beagles used in this study have been validated as suitable model for human AD which mimics natural occurring disease, with the advantage that the amount, timing, and frequency of allergen stimulation can be controlled and consecutive skin biopsies can be taken. 6 Full-thickness skin biopsies were collected from 10 atopic beagles at baseline (day 0) and at three hours (day 1, early inflammation), three days (peak inflammation), and 10 days (resolution of inflammation) following acute house dust mite (Dermatophagoides farinae) allergen stimulation. 50 mg crude D. farinae (31 mg/ml in phosphate buffered saline pH 7.2) was gently applied to the inguinal skin on days 1, 2, and 3 to provoke a moderate flare of AD which resolved by day 10. On days 1 and 3, biopsies were taken 3 h after allergen exposure. The severity of inflammation was assessed using a validated scoring system (canine atopic dermatitis extent and severity index; CADESI-03) prior to collection of each biopsy. IL-31RA splice variant transcription was measured using qPCR assays targeting total IL-31RA transcription, long IL-31RA isoforms mRNA transcript variants X1-4, short IL-31RA isoform mRNA transcript variant X5, or