U.S. Department of Agriculture: Animal and Plant Health Inspection Service


Date of this Version



Brain, Behavior, and Immunity 69 (2018) 312–320


Copyright 2017 Elsevier Inc.

This document is a U.S. government work and is not subject to copyright in the United States.



Several studies demonstrate that inflammation affects body odor. Volatile signals associated with inflammation induced by pyrogens like LPS are detectable both by conspecifics and chemical analyses. However, little is known about the mechanisms which translate detection of a foreign molecule or pathogen into a unique body odor, or even how unique that odor may be. Here, we utilized C57BL/6J trained mice to identify the odor of LPS-treated conspecifics to investigate potential pathways between LPS-induced inflammation and changes in body odor, as represented by changes in urine odor. We hypothesized that the change in volatile metabolites could be caused directly by the pro-inflammatory cytokine response mediated by TNF or IL-1b, or by the compensatory anti-inflammatory response mediated by IL-10. We found that trained biosensors generalized learned LPS-associated odors to TNF-induced odors, but not to IL-1b or IL-10-induced odors. Analyses of urine volatiles using headspace gas chromatography revealed distinct profiles of volatile compounds for each treatment. Instrumental discrimination relied on a mixture of compounds, including 2-sec-butyl-4,5-dihydrothiazole, cedrol, nonanal, benzaldehyde, acetic acid, 2- ethyl-1-hexanol, and dehydro-exo-brevicomin. Although interpretation of LDA modeling differed from behavioral testing, it does suggest that treatment with TNF, IL-1b, and LPS can be distinguished by their resultant volatile profiles. These findings indicate there is information found in body odors on the presence of specific cytokines. This result is encouraging for the future of disease diagnosis via analysis of volatiles.

Included in

Life Sciences Commons