Small mammals and their burrows shape the distribution ofCoccidioidesin soils: a long-term ecological experiment

Abstract

Background For nearly all human fungal pathogens, the environmental constraints on their distributions remain poorly understood, hindering disease management. Here, we investigated the role of zoonotic host presence, soil conditions, and their interaction on the presence of Coccidioides immitis, an emerging fungal pathogen and causative agent of coccidioidomycosis. Methods A long-term experimental study initiated in 2007 in the Carrizo Plain National Monument, California excluded rodents from certain areas (20 by 20-meter exclosures; n = 20) via the installation of submerged mesh fences, leaving emptied burrows intact. We collected 1,988 soil samples over four sampling periods spanning April 2021 to April 2022 using a factorial design that crossed burrows and surface soils (i.e., top 10 cm of soil) with exclosures (i.e., rodent absence) and non-exclosures (i.e., rodent presence). We measured in situ soil moisture, temperature, and vegetation, and analyzed soils for C. immitis DNA. Results The proportion of samples containing C. immitis was significantly higher (p <0.0001) in burrow samples taken from outside rodent exclosures (i.e., rodents present; 28.5% positive) compared to burrow samples taken from within exclosures (i.e., rodents absent; 19.7%), surface soils outside of exclosures (3.6%) and surface soils within exclosures (0.5%). In generalized linear mixed models adjusting for soil conditions, rodent presence, and sample type (burrow vs. surface soil), we estimated that the odds of detecting C. immitis were 19.2 (95% CI: 11.3 - 36.7, p < 0.0001) times higher in soils taken from burrows compared to surface soils, and 2.6 (95% CI: 1.0 - 5.6, p = 0.049) times higher in soils where rodents were present compared to absent. Soil moisture was the only abiotic factor associated with C. immitis detection (odds ratio per 1% increase in water content: 0.83, 95% CI: 0.70 - 0.89, p = 0.0002). In mediation analyses, we estimated that 73.7% (95% CI: 68.9, 78.5) of the total association between rodents and Coccidioides detection effect was attributable to rodent creation of burrows. Conclusions Rodent creation and occupation of burrows substantially increases C. immitis detection. Synergy between reservoir host presence and environmental conditions may exist for other human fungal pathogens.