House Finch (Haemorhous mexicanus) Conjunctivitis, and Mycoplasma spp. Isolated from North American Wild Birds, 1994–2015

Mycoplasma gallisepticum is a well-characterized bacterial pathogen of chickens and turkeys worldwide, and was thought to be relatively host-specific and pathogenic primarily for gallinaceous birds (Raviv and Ley 2013) until 1994, when it was identified as the cause of epidemic conjunctivitis in Eastern US House Finches (Haemorhous mexicanus; Ley et al. 1996). Disease in House Finches expanded rapidly in the Eastern North American host range (Dhondt et al. 1998) with high prevalence (Altizer et al. 2004) and high mortality (>50% population declines; Hochachka and Dhondt 2000). Dispersal to the Western US host range took several years (Duckworth et al. 2003; Ley et al. 2006) with lower prevalence and mortality (Dhondt et al. 2006). Although House Finches appeared to be the wild bird species primarily impacted by M. gallisepticum conjunctivitis, there were reports of similar disease in other wild birds, most notably American Goldfinch (Spinus tristis), Purple Finch (Haemorhous purpureus; Hartup et al. 2000), Evening Grosbeak (Coccothraustes vespertinus), and Pine Grosbeak (Pinicola enucleator; Mikaelian et al. 2001). A succession of US National Science Foundation and National Institutes of Health grants provided us the opportunity to collect samples and study this emergent disease with a multi-institutional, multidisciplinary team. This study has yielded a wealth of knowledge and has been a model of collaborative research (Dhondt et al. 2005). Key to the success and productivity of this effort has been the collection of M. gallisepticum isolates spanning the temporal and geographic ranges of the disease from emergence to endemicity in wild bird hosts. We here report compiled results of our wild bird sampling for mycoplasmas from the emergence of House Finch conjunctivitis in 1994–2015. Mycoplasma gallisepticum isolates from House Finches have been made in almost every year since 1994, for a current total of 227 isolates from 17 states (Table 1). We also found that House Finches can be infected with other Mycoplasma spp. and discovered a broader wild bird species range infected with M. gallisepticum, including this first report for American Crow and Western Scrub-jay. Also new and of interest are other mycoplasmas (primarily M. sturni) isolated from a range of wild bird species, thus expanding the previously known host-species range for this organism.

Mycoplasma culture is optimized by doing everything possible to assure organism viability with minimal delays and proper storage from collection to incubation in mycoplasma growth medium (Kleven 2008). In our case, optimal conditions were difficult to meet. The cooperators were of varied experience and from multiple locations without well-equipped laboratories; the situation was further complicated by the resulting need for sample storage and shipment to the mycoplasma laboratory. Faced with the challenges and costs (mainly overnight shipment on dry ice or cold-packs) of sample collection for mycoplasma culture from multiple locations over long periods, our sampling protocol evolved with experience. Our current mycoplasma sampling and culture protocol specifies conjunctival swabs (sterile, nylon, or polyester tips, plastic handles) inoculated to mycoplasma transport media (BD/Copan UTM, BD, Sparks, Maryland, USA or Remel M4 or M5, Remel, Lenexa, Kansas, USA) stored at 4 C and overnight shipment on cold-packs to the mycoplasma laboratory. Culture is initiated immediately upon arrival with the use of Frey's medium with 15% swine serum incubated at 37 C (Kleven 2008). Even with this protocol, mycoplasma culture-positive rates are highly variable among samples submitted, but can attain 50–90% from House Finches with conjunctivitis.

Wild bird mycoplasma isolation and identification results since 1994 are summarized in Table 1. Most of these are Mycoplasma spp. culture isolates (now in archival storage at −70 C) with identification of M. gallisepticum and M. sturni by species-specific immunofluorescence of colonies (Ley et al. 1998; Kleven 2008). Mycoplasma gallisepticum–specific PCR (Garcia et al. 2005) and 16S rRNA gene sequencing (Ley et al. 2012) were also used in some cases for Mycoplasma spp. identification. The majority of our isolates are from House Finches, the original wild bird host of interest. We have M. gallisepticum isolates from House Finches almost every year since disease emergence in 1994 from Virginia, and now from 17 states for a total of 227 isolates. A 2015 House Finch sample from Colorado was M. gallisepticum PCR–positive, but culture was not successful. Surprisingly, we also isolated two other Mycoplasma spp. from House Finches. In 2006, we identified a House Finch isolate made in California as M. sturni that was not pathogenic by experimental infection (Ley et al. 2010) and remains a unique finding. In a 2015 California submission from three House Finches and one Red-tailed Hawk (Buteo jamaicensis), M. gallisepticum was isolated from a House Finch and M. gypis was isolated from the Red-tailed Hawk and the other two House Finches, a novel finding in this host. We also have unidentified isolates from California: three from House Finches, three from American Crows (Corvus brachyrhynchos), and one from a Red-shouldered Hawk (Buteo lineatus).

In addition to isolating multiple Mycoplasma spp. from House Finches, we have isolated M. gallisepticum and M. sturni from diverse wild bird species. Our M. gallisepticum host range includes House Finch, Blue Jay (Cyanocitta cristata), American Goldfinch, Lesser Goldfinch (Spinus psaltria), Purple Finch, Evening Grosbeak, Western Scrub-jay (Aphelocoma californica), American Crow, and Black-capped Chickadee (Poecile atricapillus; by molecular identification, not cultured). Our M. sturni host range includes House Finch, Blue Jay, Northern Mockingbird (Mimus polyglottos), European Starling (Sturnus vulgaris), American Crow, American Robin (Turdus migratorius), Carolina Wren (Thryothorus ludovicianus), Cliff Swallow (Petrochelidon pyrrhonota), and Barn Swallow (Hirundo rustica; by molecular identification, not cultured).

Selected M. gallisepticum isolates in experimental infections of House Finches showed evolution of virulence with parallel patterns of increased virulence in both Western and Eastern isolates (Hawley et al. 2013). Phylogenetic studies with our M. gallisepticum isolates from wild birds and poultry suggested at least two host transfers from poultry to house finches, but only one successful lineage accounting for the continent-spanning epidemic (Hochachka et al. 2013). Mycoplasma gallisepticum isolates from our collection showed extensive variation in surface lipoprotein gene content, phenotypic plasticity, and genomic changes (Tulman et al. 2012).

Accumulating evidence shows that the wild bird host range for M. gallisepticum is surprisingly diverse. With the use of serology and PCR, we recently found that a diverse range of wild bird species may carry or have been exposed to M. gallisepticum in the US (Dhondt et al. 2014), as in Europe (Pennycott et al. 2005) and Asia (Shimizu et al. 1979; Ganapathy et al. 2007). Infections of the broader host range of wild birds could represent further host switching by the House Finch clade or multiple lineages of M. gallisepticum. Additional work is needed to identify the phylogenetic relationships of M. gallisepticum strain(s) infecting the entire array of wild bird host species. It is possible that M. gallisepticum detected in wild birds globally represents past or recent introductions from poultry, with noncommercials (backyard) poultry being the most common reservoirs of diverse M. gallisepticum strains (McBride et al. 1991; Ewing et al. 1996; Thekisoe et al. 2003) to interface with wild birds.

Mycoplasmas are more often commensals than pathogens, causing subclinical and chronic or latent infections (Bradbury 2005; Citti and Blanchard 2013). Evidence for a diverse wild bird host range infected with M. gallisepticum may be another example of transmissible subclinical mycoplasmosis, achievement of an ideal host/parasite relationship. Emergence of a pathogenic M. gallisepticum strain in House Finches may be the exception that has allowed us to identify the broader epidemiologic picture.

We thank Sile Huyan and Judith McLaren for mycoplasma culture and identification technical assistance. For sample submissions we thank personnel at The Wildlife Center of Virginia, Cornell Lab of Ornithology, Virginia Tech Department of Biological Sciences, University of Georgia Institute of Ecology, Audubon Society of Portland Wildlife Care Center, California Wildlife Center, Lindsay Wildlife Museum, North Carolina Zoological Park Valerie H. Schindler Wildlife Rehabilitation Center, Greenwood Wildlife Rehabilitation Center, Piedmont Wildlife Center, and Triangle Wildlife Rehabilitation Clinic. This work was supported by National Science Foundation (NSF) grants DEB 0094456 and EF 0622705 to A.D.D. as part of the National Institutes of Health (NIH)–NSF Ecology of Infectious Diseases program; and NSF grant 0731894 to S.J.G.; and to D.M.H., NIH grant 5R01GM105245, as part of the joint NIH-NSF-US Department of Agriculture Ecology and Evolution of Infectious Diseases program.