Response of Black-Capped Chickadees to House Finch Mycoplasma gallisepticum

Tests for the presence of pathogen DNA or antibodies are routinely used to survey for current or past infections. In diseases that emerge following a host jump estimates of infection rate might be under-or overestimated. We here examine whether observed rates of infection are biased for a non-focal host species in a model system. The bacterium Mycoplasma gallisepticum is a widespread pathogen in house finches (Haemorhous mexicanus), a fringillid finch, but an unknown proportion of individuals of other songbird species are also infected. Our goal is to determine the extent to which detection of M. gallisepticum DNA or antibodies against the bacteria in a non-fringillid bird species is over-or underestimated using black-capped chickadees Poecile atricapillus, a species in which antibodies against M. gallisepticum are frequently detected in free-living individuals. After keeping black-capped chickadees in captivity for 12 weeks, during which period the birds remained negative for M. gallisepticum, four were inoculated with M. gallisepticum and four were sham inoculated in both eyes to serve as negative controls. Simultaneously we inoculated six house finches with the same isolate of M. gallisepticum as a positive control. All inoculated birds of both species developed infections detectable by qPCR in the conjunctiva. For the 6 weeks following inoculation we detected antibodies in all M. gallisepticum-inoculated house finches but in only three of the four M. gallisepticum-inoculated black-capped chickadees. All house finches developed severe eye lesions but none of the black-capped chickadees did. Modeling the Rapid Plate Agglutination test results of black-capped chickadees shows that the rate of false-positive tests would be not more than 3.2%, while the estimated rate of false negatives is 55%. We conclude that the proportion of wild-caught individuals in which we detect M. gallisepticum-specific antibodies using Rapid Plate Agglutination is, if anything, substantially underestimated.