Coxiella burnetii in Wild-caught Filth Flies

To the Editor: Coxiella burnetii, the agent of Q fever, is a bacterium and a potential agent of bioterrorism. The most frequent signs of infection in domestic animals are abortion and reduced fertility (1). Clinical signs of Q fever in humans vary from mild fevers to pneumonia, hepatitis, or death; atypical cases occur as other disorders, such as cholecystitis (1,2). Aerosols are the most common route of exposure, but oral transmission occurs (1). 
 
Some flies feed on the feces, milk, carcasses, or blood of domestic animals that can be infected with C. burnetii. These flies regurgitate and defecate when feeding and are mechanical vectors of bacteria (3,4). Flies have been shown to harbor, mechanically transport, and even support the growth of C. burnetii (4–6). It is known that house flies (Musca domestica) are possible mechanical vectors of C. burnetii because this organism survived 32 days in house flies and viable bacteria were shed by flies for 15 days (4). There are no studies of C. burnetii in field-collected flies. To examine the prevalence of C. burnetii in field-collected flies, we tested flies from farms, forests, ranches, and zoos. 
 
Flies that develop on animal dung, carcasses, feces, blood, or garbage are often called filth flies. Adult Calliphoridae, Hippoboscidae, Muscidae, and Sarcophagidae were collected from forests, zoos, ranches, and farms (Table). Flies were killed in 95% ethanol or by freezing. DNA was extracted from individual flies as described (7,8). A distilled water negative control was used for each extraction. 
 
 
 
Table 
 
Flies from the United States and Dominica assayed for Coxiella burnetii, 2004–2007 
 
 
 
Individual DNA samples were tested, in duplicate, with a previously described TaqMan assay with a lower limit of detection of 1 C. burnetii organism (8). Positive and negative controls were used for all assays. Positive flies were verified by PCR and sequencing of 16S rRNA gene as described (9). Vouchers for each insect species were deposited in the Clemson University Arthropod Collection (Clemson, SC, USA), the University of Georgia Museum of Natural History (Athens, GA, USA), or the University of Wyoming Insect Collection (Laramie, WY, USA). 
 
Five of 363 flies were positive for C. burnetii DNA (Table). These flies included Stomoxys calcitrans, in which the adults feed on animal and human blood, and the blowflies Lucilia coeruleiviridis and L. sericata. C. burnetii–positive flies were obtained from carrion (1/12, 8.3%), a garbage bin of elephant feces (3/18, 16.7%), and a barn at a ranch (1/55, 1.8%). We sequenced 1,100 bp of the 16S rRNA gene from select DNA extracts, which were 99% identical with that of C. burnetii strain NC 002971. 
 
We detected DNA from C. burnetii in flies from a zoo, a ranch, and carrion in a forest. Laboratory data on house flies, which shed live C. burnetii for 15 days after exposure, suggest that related flies (e.g., S. calcitrans and Lucilia spp.) might also harbor viable C. burnetii. On the basis of our field data, S. calcitrans and Lucilia spp. should be studied as mechanical vectors of C. burnetii. Unlike many enteric bacteria, which require large inocula to cause disease, C. burnetii can be infectious at the level of 1 bacterium (10). If flies transmit C. burnetii, they pose an additional threat to human and animal health. 
 
The role of the sheep ked (Melophagus ovinus) in maintenance or transmission of C. burnetii is unknown. This fly is an obligate ectoparasite of sheep. It feeds on sheep blood, and feces from sheep keds can accumulate in the wool of sheep. Testing of sheep keds from infected sheep would help understand whether keds play a role in the epidemiology of C. burnetii.

human mitochondrial DNA HVS-I region characterized this person as belonging to haplogroup B (GenBank accession no. EU359272), one of the founder human haplogroups in the Americas.
The antiquity of human T. cruzi infection in South America has been demonstrated on the basis of paleonthologic studies. Clinical manifestations of Chagas disease were observed in Chilean mummies from pre-Columbian times (7). Moreover, a T. cruzi kinetoplast DNA region was recovered in Chilean and Peruvian mummies from up to 9,000 years ago (8,9).
In Brazil, the current epidemiologic scenario concerning Chagas disease in indigenous populations involves ecologic aspects of their settlements, along with nomad habits, which prevent triatomine nesting and, therefore, the infection. The beginning of T. cruzi transmission to humans is attributed to the domiciliation of T. infestans as a consequence of precarious mud dwellings, built after European colonization (10). In this report, we showed that T. cruzi human infection in Brazil is ancient, dating back at least 4,500 years, and therefore occurring in hunter-gatherer populations largely preceding T. infestans domiciliation. The presence of the T. cruzi I genotype infecting humans 4,500-7,000 years ago in Minas Gerais State, where this genotype is currently absent (6), suggests that the distribution pattern of T. cruzi genotypes in humans has changed in time and place. Moreover, the recovery of an aDNA sequence and the possibility of genotyping parasites from human remains make it possible to reconstruct the early dispersion patterns of T. cruzi subpopulations. On the basis of our results, one may speculate that the current outbreaks of human T. cruzi infection, independent of triatomine domiciliation, are the reemergence of the ancient epidemiologic scenario of Chagas disease in Brazil.

Coxiella burnetii in Wild-caught Filth Flies
To the Editor: Coxiella burnetii, the agent of Q fever, is a bacterium and a potential agent of bioterrorism. The most frequent signs of infection in domestic animals are abortion and reduced fertility (1). Clinical signs of Q fever in humans vary from mild fevers to pneumonia, hepatitis, or death; atypical cases occur as other disorders, such as cholecystitis (1,2). Aerosols are the most common route of exposure, but oral transmission occurs (1).
Some fl ies feed on the feces, milk, carcasses, or blood of domestic animals that can be infected with C. burnetii. These fl ies regurgitate and defecate when feeding and are mechanical vectors of bacteria (3,4). Flies have been shown to harbor, mechanically transport, and even support the growth of C. burnetii (4)(5)(6). It is known that house fl ies (Musca domestica) are possible mechanical vectors of C. burnetii because this organism survived 32 days in house fl ies and viable bacteria were shed by fl ies for 15 days (4). There are no studies of C. burnetii in fi eld-collected fl ies. To examine the prevalence of C. burnetii in fi eld-collected fl ies, we tested fl ies from farms, forests, ranches, and zoos.
Flies that develop on animal dung, carcasses, feces, blood, or garbage are often called fi lth fl ies. Adult Calliphoridae, Hippoboscidae, Muscidae, and Sarcophagidae were collected from forests, zoos, ranches, and farms (Table). Flies were killed in 95% ethanol or by freezing. DNA was extracted from individual fl ies as described (7,8). A distilled water negative control was used for each extraction.
Individual DNA samples were tested, in duplicate, with a previously described TaqMan assay with a lower limit of detection of 1 C. burnetii organism (8). Positive and negative controls were used for all assays. Positive fl ies were verifi ed by PCR and sequencing of 16S rRNA gene as described (9). Vouchers for each insect species were deposited in the Clemson University Arthropod Collection (Clemson, SC, USA), the University of Georgia Museum of Natural History (Athens, GA, USA), or the Uni-versity of Wyoming Insect Collection (Laramie, WY, USA).
Five of 363 fl ies were positive for C. burnetii DNA (Table). These fl ies included Stomoxys calcitrans, in which the adults feed on animal and human blood, and the blowfl ies Lucilia coeruleiviridis and L. sericata. C. burnetii-positive fl ies were obtained from carrion (1/12, 8.3%), a garbage bin of elephant feces (3/18, 16.7%), and a barn at a ranch (1/55, 1.8%). We sequenced 1,100 bp of the 16S rRNA gene from select DNA extracts, which were 99% identical with that of C. burnetii strain NC 002971.
We detected DNA from C. burnetii in fl ies from a zoo, a ranch, and carrion in a forest. Laboratory data on house fl ies, which shed live C. burnetii for 15 days after exposure, suggest that related fl ies (e.g., S. calcitrans and Lu-cilia spp.) might also harbor viable C. burnetii. On the basis of our fi eld data, S. calcitrans and Lucilia spp. should be studied as mechanical vectors of C. burnetii. Unlike many enteric bacteria, which require large inocula to cause disease, C. burnetii can be infectious at the level of 1 bacterium (10). If fl ies transmit C. burnetii, they pose an additional threat to human and animal health.
The role of the sheep ked (Melophagus ovinus) in maintenance or transmission of C. burnetii is unknown. This fl y is an obligate ectoparasite of sheep. It feeds on sheep blood, and feces from sheep keds can accumulate in the wool of sheep. Testing of sheep keds from infected sheep would help understand whether keds play a role in the epidemiology of C. burnetii. Cholera, caused by the bacterium Vibrio cholerae, is a disease that seems particularly sensitive to confl ict and deserves more consideration. Major risk factors for cholera-poverty, overcrowding, poor hygiene, contaminated food, and lack of safe drinking water (2,3)-largely resemble the consequences of war and civil fi ghting. Yet little is known about the relationship between cholera and confl ict. This lack of information may be because cholera tends to be epidemic, affecting hundreds to thousands of people across vast, war-torn regions, making it impossible for local governments, nongovernment organizations, and aid workers to control, let alone collect and analyze data.
Examination of data sources listed by Gayer et al. (1) and recent reviews (2,3) indicate that cholera occurs 1) in countries during war and civil unrest, as exemplifi ed by the latest outbreaks among displaced populations across northern Iraq; 2) in neighboring countries, where temporary camps accommodate masses of political refugees under poor conditions, such as those in eastern Chad near Darfur, Sudan; and 3) during the postwar period when large numbers of repatriated persons return home and consequently place undue pressure on an eroded and fragile national infrastructure, as evident in Angola in recent years.
Moreover, all the countries affected by confl ict shown in the Figure by cdc.gov/EID/content/13/11/1625-G. htm) have reported cholera outbreaks (2)(3)(4). They are also among the poorest countries in the world; the latest statistics on human development (5) indicate that compared with all developing countries, on average they have higher rates of undernourishment, refugees, child deaths, and less adequate water and sanitation facilities. Thus, more information is needed about confl ict and cholera, especially in Africa.
Louise A. Kelly-Hope* *Liverpool School of Tropical Medicine, Liverpool, UK