Exposure levels of animal allergens, endotoxin, and β-(1,3)-glucan on a university campus of veterinary medicine

Abstract

Objectives: The study aimed to determine the allergen, endotoxin and β-(1,3)-glucan concentrations at various areas on a university campus of veterinary medicine.

Methods: Dust samples were collected four times a year for three years using electrostatic dust collectors (EDC) at 25 different locations on a campus of veterinary medicine and in laboratories of inorganic chemistry as a control area representing animal-free environment. Major animal allergens from dog, cat, horse, cattle and mouse, domestic mite (DM) allergens, and β-(1,3)-glucan were measured using enzyme immunoassays and endotoxin using the limulus amoebocyte lysate (LAL) assay. Seasonal, annual and local influences on exposure levels were analyzed using Bayesian mixed models.

Results: With the exception of mouse allergens, all other determinants were found in almost all locations on the campus and in the control area, but in up to 10.000-fold variable concentrations. By far the highest levels of feline, canine, equine and bovine allergens were detected in buildings where the respective species were examined. The highest levels of mouse and DM allergens, β-(1,3)-glucan and endotoxin occurred together and were associated with locations where large animals were present. In buildings without animals, allergen levels were considerably lower but still elevated at several locations compared to the control area, especially for dog and horse allergens, and β-(1,3)-glucan. Significant seasonal effects were observed for dog, cat, horse and DM allergens, and β-(1,3)-glucan. Variations between years were less apparent than between seasons (except for β-(1,3)-glucan).

Conclusions: The strongest influencing factor on the concentration of mammalian allergens was the presence of the corresponding animal at the collection site. Seasonal influence on allergen concentrations was observed, while the overall exposure remained constant over the years. At locations with horses, elevated levels of mite allergens, endotoxin, and β-(1,3)-glucan can be expected, probably due to passive transfer from stable environment.

Fig 1

Seasonal and annual variations of concentrations of A) cat allergen, B) dog allergen, C) horse allergen, D) cattle allergen, E) domestic mite allergens, F) mouse allergen (only descriptive measures), G) β-(1,3)-glucan and H) endotoxin. Following parameters of the Bayesian models are shown: the point estimate—of the posteriors—as an index of centrality (red circle) and the 95% credible interval characterizing the dispersion of the posteriors (whiskers). The results of Bayesian hypothesis tests are marked with asterisks: ***: pd>99.9% (certainly existing) & <1% in ROPE (significant), **: pd>99.9% (certainly existing) & <2.5% in ROPE (probably significant) OR pd>99% (probably existing) & <1% in ROPE (significant), *: pd>99% (probably existing) & <2.5% in ROPE (probably significant), The exact pd and ROPE values are listed in S3 Table.

Fig 2. Heatmap of the Spearman correlation matrix of all bio-contaminant concentrations.