Mycobacterium bovis Shuttles between Domestic Animals and Wildlife

Mycobacterium bovis Shuttles between Domestic Animals and Wildlife This infectious agent that once spilled over from cattle to wildlife now mainly moves the other way, from wildlife to cattle Mitchell V. Palmer Summary Mycobacterium bovis has an extremely broad host range that includes most mammals, sometimes leading to localized outbreaks that become endemic. White-tailed deer in Michigan are the first known U.S. reservoir of M. bovis in free-living wildlife and the first known epizootic of tuberculosis involving this species. Although M. bovis was likely introduced to New Zealand with imported cattle during the early 19th century, the first case of tuberculosis in a wild possum was not recorded there until 1967. U.K. authorities believe that badgers became infected with M. bovis from cattle during the late 19th century, and now serve as a source for this pathogen. Because M. bovis is maintained in wildlife populations and transmitted to livestock, animal health authorities face major challenges when developing tuberculosis eradication programs. nfectious disease outbreaks affecting I wildlife sometimes involve complex, unintended interactions among wildlife, domestic animals, and humans. True emerging diseases in wildlife include house finch conjunctivitis, amphibian chytridiomycosis, and chronic wasting disease. However, many emerging diseases such as tuberculosis in wildlife species are really diseases that spill over from domestic animals, according to Peter Daszak of the University of Georgia, who cites factors such as human encroachment on traditional wildlife habitat, agricultural intensification, introductions of wildlife species, and disease spillovers. Tuberculosis illustrates many key themes associated with diseases at the interface of domestic animals and wildlife. The bacterial pathogen that is responsible, Mycobacterium bovis, has an extremely broad host range that includes most mammals, as well as humans, causing symptoms that are virtually indistinguishable from those caused by M. tuberculosis, the cause of most human cases of tuberculosis. Although human cases of M. bovis infection are uncommon in developed countries, they still represent a serious public health concern in underdeveloped agrarian societies. In the United States, M. bovis was responsible for approximately 25% of human tuberculosis cases prior to mandatory pasteurization of milk and disease control programs that began in 1917, reducing the prevalence of disease in cattle from 5% to 0.0002%. However, one serious obstacle to eradication became apparent in 1995, when M. bovis infection was identified in white-tailed deer, Odocoileus virginianus, in northeast Michigan. This same spillover with subsequent spillback of tuberculosis is a theme repeated in several other parts of the world attempting to eradicate M. bovis from domestic herds. Spillover and Spillback Before 1994, reports of tuberculosis in hunter-killed or accidentally killed whitetailed U.S. deer indicated that M. bovis had spilled over from livestock. For example, although Michigan farmers for several de- Mitchell Palmer is a scientist and veterinary medical officer in the Bacterial Diseases of Livestock Research Unit, at USDA s National Animal Disease Center in Ames, Iowa. Volume 3, Number 1, 2008 / Microbe Y 27

FIGURE 1 Winter feeding: (top) Large numbers of deer congregating around corn provided as supplemental feed. (Bottom) Aerial view of large winter feed site. Numerous deer trails demonstrate the large numbers of deer that visit site. (Photos courtesy of Michigan Department of Natural Resources.) cades were reporting large numbers of tuberculous cattle, the diagnosis of M. bovis tuberculosis in a free-ranging white-tailed deer in northern Michigan in 1975 was considered only an anomaly. That same year, U.S. Department of Agriculture (USDA) officials declared Michigan livestock free of M. bovis and subsequently granted the state TB-free status in 1979. However, in 1994 a free-ranging white-tailed deer was the identified with M. bovis tuberculosis just 13 km from where the tuberculous deer was identified in 1975. Subsequently, investigators from the Michigan Department of Natural Resources and the Michigan State University (MSU) Animal Health Diagnostic Laboratory identified a focus of M. bovis infection in free-ranging white-tailed deer in northeast the Michigan. These animals made up the first known U.S. reservoir of M. bovis in free-living wildlife and the first known epizootic of tuberculosis in white-tailed deer in the world. Several factors contributed to this M. bovis wildlife reservoir. Initially, M. bovis likely was transmitted from cattle to deer during the early to mid- 20th century when large numbers of Michigan cattle were infected with M. bovis. Spillover from cattle to deer occurred around 1955, according to estimates by C. W. McCarty from Colorado State University and Michael Miller from the Colorado Division of Wildlife. During that same period, Michigan s deer population steadily increased beyond the habitat s usual carrying capacity. Thus, deer in Michigan increased from about 592,000 in 1930 to more than 1.7 million by 1998, with focal concentrations of 19 to 23 deer per square kilometer. Regions of highest deer density were at the center of the tuberculosis outbreak, according to Steve Schmitt and Dan O Brien, wildlife veterinarians from the Michigan Department of Natural Resources. Similar explosive growth in whitetailed deer populations was also seen in other regions. Overall, U.S. deer populations grew from fewer than 300,000 in 1900 to an estimated 19 million by 1996. The Dangers of Supplemental Feeding of Wildlife Michigan residents help to maintain M. bovis among deer herds by setting out large volumes of sugar beets, carrots, corn, apples, pumpkins, and pelleted feed during the winter to help keep numbers high for hunting season (Fig. 1). The resultant crowding around feeding sites provides plenty of opportunity for deer-to-deer con- 28 Y Microbe / Volume 3, Number 1, 2008

Palmer s Interests Led from Veterinary Medicine to Research on Zoonoses When Mitch Palmer graduated from veterinary school nearly 20 years ago, his grandfather presented him with a treasured volume that had served him well working with animals. He hoped it would similarly inspire his grandson. In Western Utah, where my family came from, my grandfather, with an 8th-grade education, was the closest thing to a veterinarian that most people in the area knew, Palmer says. He was a rancher, farrier, miner, and railroad man who worked from a book, entitled Veterinary Science, published in 1905. He castrated calves, pigs, sheep, and horses, and treated sick pigs, calves, and lame horses. I would say he had a profound influence on me, and when I graduated from Purdue College of Veterinary Medicine in 1989, he passed along his 1905 text to me. Palmer, 47, who was born in Salt Lake City and raised in Utah, was interested in agriculture from childhood. This interest persisted, leading him as a young man into veterinary science, a practice specializing in large animals, and, later, to conduct research on diseases that are transmitted between wildlife and livestock. Although his parents did not farm his father is a computer scientist, his mother a medical records consultant his extended family was involved with agriculture, and Palmer performed his share of farm work from an early age. When he was active as a youngster in FFA, formerly Future Farmers of America, my projects involved raising Suffolk sheep, he says. Later, he became fascinated with biology and physiology. I loved to know how things worked in the body, he says. I still remember seeing a movie in biology class called Man the Magnificent Machine, and being awestruck at how the different body systems worked together in harmony to keep us alive. Veterinary medicine seemed to be the perfect union of my interest in science and agriculture. Today Palmer is a collaborating assistant professor in the veterinary pathology department at Iowa State University in Ames and a veterinary medical officer in the bovine tuberculosis laboratory at the nearby U.S. Department of Agriculture facility specializing in bacterial diseases of livestock. He is particularly interested in zoonotic diseases, those that can afflict humans as well as other animal species, or that move between wild and domestic animal species. Most of my research has focused on studying the progression of disease in the host, how the disease is transmitted between two populations, methods to diagnose the disease, and vaccines to prevent the disease in either the wildlife or domestic animal population, he says. Palmer received a Bachelor of Science degree in veterinary science from Utah State University in 1985, and a doctorate degree in veterinary medicine from Purdue University in 1989. He spent several years working in a large animal practice and enjoyed it before deciding to switch to research. I convinced my understanding wife and four daughters that I needed to obtain a Ph.D., he says. They moved to Ames where, in 1996, he received his doctorate in veterinary pathology from Iowa State University and began working at the USDA National Animal Disease Center. Palmer met his wife, also from Utah, in Spain where they were doing missionary work. They have been married almost 25 years. In addition to four daughters, they have a granddaughter whose image is on my computer desktop, he says. When his daughters were very young, they sometimes helped him with emergency cases. They especially liked it when I performed C- sections on dogs, he recalls. As I took each puppy out, they would take it in a warm towel, dry it off, make sure it was breathing, and put it in an incubator. Once I was bottle-feeding a newborn deer fawn. I brought it home since it had to be fed every few hours. They had great fun taking turns feeding it. Palmer s wife, formerly an elementary school teacher, runs a daycare center. I think she has raised half the kids in our community, but it has allowed her to be home with our children, and now our granddaughter, which was important to us, he says. Both being small-town people, they live outside a small community of 5,000 people in central Iowa. Although we both miss the mountains of Utah, we have found Iowa a great place to raise a family, he says. In his spare time, Palmer enjoys jogging, photography particularly in southern Utah, with its red rock natural architecture and pheasant hunting with the only other male in our family, a threeyear-old Brittany spaniel, he says. Photography is my chance to exercise my left brain, and I especially like to photograph nature from unusual perspectives. Marlene Cimons Marlene Cimons is a freelance writer in Bethesda, Md. Volume 3, Number 1, 2008 / Microbe Y 29

FIGURE 2 Medial retropharyngeal lymph node from a deer with tuberculosis. Lesions similar to this can be easily mistaken for abscesses. (Photo by Mitch Palmer.) tact, enhancing M. bovis transmission. Additional specific risk factors associated with increased risk of tuberculosis in deer include locating feeding sites near hardwood forests, numbers of deer fed per year, presence of other nearby feeding sites, and the quantity of grain, fruits, or vegetables fed, according to Rose Ann Miller of MSU in East Lansing. Most Michigan white-tailed deer are infected with a common strain of M. bovis, suggesting a single source of infection, according to restriction fragment length polymorphism (RFLP) analysis. By mid-2006, more than 145,000 deer had been tested since the first case was found in 1994. Of these, 525 confirmed M. bovis cases were identified in 12 counties in northern Michigan. M. bovis in wildlife not only is detrimental to those animals but also seriously affects livestock, with at least 40 M. bovis-infected cattle herds identified in Michigan since 1994. M. bovis from deer and cattle are identical, according to RFLP analysis, suggesting free-ranging deer infected cattle there. These findings cost Michigan its USDA TB-free status, although portions of the state subsequently were again declared TB-free. This status and the continued presence of tuberculosis in cattle has important economic consequences in both domestic and international markets, with livestock producers also bearing the additional burden of increased testing and occasional losses of herds to control this disease. White-tailed deer experimentally infected with M. bovis shed bacilli in saliva and nasal secretions, and less frequently in urine and feces, in experiments that we directed at the USDA National Animal Disease Center in Ames, Iowa. These deer can transmit M. bovis to other deer or cattle through indirect contact such as sharing of feed. Saliva and nasal secretions containing M. bovis can contaminate feed to become a source of infection. M. bovis is relatively resistant to environmental factors and may persist for weeks or months (Fig. 1). Dan O Brien suggests a two-stage model of transmission, initially allowing disease to persist at a low level within matriarchal groups composed of a doe and several generations of her daughters and fawns. Stage 2 involves both supplemental feeding, with resultant increased deer density, and male fawns that disperse to join male groups that travel together at all times except during breeding season. Higher disease prevalence has been observed in adult male deer. Shifting membership by many males results in males temporarily belonging to several different groups and increased contact with numerous susceptible animals. Naturally infected white-tailed deer commonly develop lesions of tuberculosis in retropharyngeal lymph nodes and, less commonly, in the lung and pulmonary lymph nodes (Fig. 2), according to Scott Fitzgerald of MSU. Similar to other species of deer, lesions resemble abscesses, making differential diagnosis important. Microscopically, lesions consist of foci of necrosis, surrounded by infiltrates of epithelioid macrophages, lymphocytes, and Langhan s multinucleated giant cells. Lesions are often surrounded by fibrous connective tissue, while acid-fast bacilli may be present within the caseum, macrophages, or multinucleated giant cells (Fig. 3). Although M. bovis is a zoonotic agent, the 30 Y Microbe / Volume 3, Number 1, 2008

incidence of M. bovis infections in humans has remained constant in Michigan throughout this epizootic. However, two clinical cases are linked to the M. bovis strain in free-range deer. In one case, a hunter developed cutaneous tuberculosis after injuring himself while field-dressing an infected whitetailed deer (Fig. 4). More generally, hunters are exposed to M. bovis when field-dressing deer or consuming undercooked venison. The Michigan Departments of Community Health, Natural Resources, and Agriculture continue to educate hunters, farmers, and other state residents on how to identify tuberculosis in deer, to take protective measures when dressing deer carcasses, and to cook venison thoroughly. FIGURE 3 Nonnative Possums in New Zealand Transmit M. bovis to Cattle European settlers introduced cattle into New Zealand two centuries ago, and soon cleared large tracts of forests to accommodate the new herds. Meanwhile, seafarers began bringing brushtail possums (Trichosurus vulpecula) to New Zealand from Australia during the mid-19th century, eventually releasing possums at more than 160 sites. A lack of natural predators and access to abundant food led possum populations to grow explosively, enabling them to occupy more than 90% of the New Zealand land area and to build a population size estimated at 60 70 million. M. bovis was likely introduced to New Zealand along with imported cattle during the early 19th century. However, the first case of tuberculosis in a wild possum was not recorded until 1967 by an official of the New Zealand Department of Agriculture. Since then, growing epidemiological evidence links possum tuberculosis to tuberculosis among cattle thus providing another example of spillover from cattle to possums. Possums infected with M. bovis often develop disseminated disease, with lymph nodes and (A) Tuberculous granuloma with necrotic core, central mineralization (brown staining), inflammatory cell rich zone (between black arrows) and numerous Langhan s type multinucleated giant cells (white arrows). (B) Acid-fast M. bovis bacilli within the necrotic core of a granuloma. (Photos by Mitch Palmer.) lungs being the most common sites of infection. At least 45% of affected possums develop discharging sinuses, according to Michele Cooke of Massey University in New Zealand. Lesions also occur in the liver, spleen, kidney, adrenal gland, and bone marrow of infected possums, suggesting hematogenous spread of bacilli. Numerous organ systems are affected in terminally ill possums, resulting in profound effects on behavior. All in all, possums efficiently maintain this Volume 3, Number 1, 2008 / Microbe Y 31

FIGURE 4 of tuberculosis. However, various control measures, including a bounty on possums, proved only minimally effective in part, because those programs focused on possums in easily accessible locations. When officials distributed baits containing poison to kill possums, tuberculosis rates dropped but eventually returned to high levels. Although widespread reductions of possums through poisoning may decrease the prevalence of tuberculosis in cattle, removing possums from New Zealand may prove impractical. Perhaps a more promising option for long-term control of tuberculosis would be to develop and use a vaccine along with biological control of the possum population. Internal thoracic cavity of a deer with severe miliary tuberculosis. Field dressing of such a deer by a hunter could pose risk of deer-to-human transmission. (Photo courtesy of Michigan Department of Natural Resources.) pathogen and transmit it to other susceptible individuals. Respiratory secretions are the main means by which the pathogen is transmitted from one possum to another possum, according to Ron Jackson of Massey University. However, pathogens also are transmitted via draining lesions as well as by milk from mothers to offspring. Healthy possums generally avoid contact with cattle; however, terminally ill possums exhibit abnormal behaviors in which cattle take a profound interest, according to Roger Morris and colleagues of Massey University. Public health officials and many environmentalists in New Zealand view possums as nonnative, invasive pests whose removal is desirable for many reasons in addition to tuberculosis control. For example, the 60-70 million possums continue to damage native flora and fauna, daily consuming approximately 21,000 metric tons of green shoots, leaves, and berries. Possums also consume eggs, chicks, and insects, while competing with nectar-feeding birds as well as with native kiwi for dens. In theory, removing nonnative possums from New Zealand should be more palatable than removing native wildlife that serve as reservoirs U.K. Badgers Are Reservoir Species for Tuberculosis Although animal health authorities saw the prevalence of bovine tuberculosis drop throughout much of Great Britain during the 1970s, its incidence in cattle was again on the rise during the past decade in England, South Wales, and the Republic of Ireland. One likely reason for this resurgence is that M. bovis is endemic among badgers (Meles meles), which are considered a probable source of infection for cattle. Authorities believe that badgers became infected with M. bovis during the late 19th and early 20th centuries from cattle, which at the time were widely infected with this pathogen. Legislation passed several decades ago, including the Badgers Act and the Wildlife and Countryside Act, protects U.K. badgers and led to a large increase in their numbers. Badgers live on or near pastures, where they live in and defend communal territories that typically include several setts, which consist of complex networks of tunnels and channels that incidentally provide ideal conditions for spreading respiratory diseases. However, badger social groups, which may remain stable for years, tend not to disperse and thus not to transmit diseases between such groups. The lesions that tuberculous badgers develop differ from those typically seen in cattle, and thus have important implications in disease pathogenesis and transmission. In particular, 32 Y Microbe / Volume 3, Number 1, 2008

the Langhan s-type giant cells as well as the caseous necrosis, mineralization, and peripheral fibrosis that typifies tuberculous lesions in cattle are rare among badgers. Conversely, acid-fast bacilli, which are present in low numbers in bovine lesions, are often numerous in lesions of diseased badgers. Renal lesions are more common in badgers than in cattle. Badgers can transmit M. bovis to cattle by several routes, according to Thomas Little and collaborators at the Central Veterinary Laboratory in Weybridge, U.K. In the wild, badgers shed large numbers of M. bovis in saliva, urine, feces, and exudates from draining lesions, making it likely that cattle become infected by inhaling bacilli from contaminated grasses. Moreover, because badgers live 3 4 years after becoming infected, they are an effective host for harboring M. bovis. Indeed, areas of high badger density have the highest incidence of tuberculosis among cattle. Badger-to-badger transmission is through aerosols and wounds. Recently, public health officials documented several cases of bovine tuberculosis spilling over from animals to humans. Two siblings residing on a farm were diagnosed with M. bovis tuberculosis, as were cattle on that farm. Isolates from cattle on the farm were indistinguishable from the human isolates by several criteria, including RFLP and variable number tandem repeat (VNTR) analyses, and thus consistent with the infection being transmitted between cattle and humans. Moreover, the farm supported a large badger population in which M. bovis infection was detected. Although not proved, the humans apparently were infected by cattle that were infected through contact with badgers. Efforts To Cull or Vaccinate against M. bovis Prove Frustrating Although reducing badger populations in cattle farming areas led to a decline in bovine tuberculosis, the effectiveness of large-scale culling is under debate. In 1998, English authorities began a large experiment to compare the effects of three tuberculosis control strategies: no culling of badgers, localized or reactive culling in response to tuberculosis in cattle, and proactive culling to reduce badger densities across specific areas. Surprisingly and inexplicably, reactive culling of badgers led to increased levels of tuberculosis in cattle within trial areas. However, because badger social structures are complex, selectively removing some but not all badgers may increase movement by the remaining population, leading to enlarged social groups with overlapping boundaries. Such social restructuring among populations of M. bovis-infected badgers may transmit more infections among badgers and between badgers and cattle. Increased social restructuring and badger movement is correlated with increased incidence of M. bovis infection among badger populations. Meanwhile, proactive culling of badgers reduced cattle tuberculosis rates within trial areas. However, those localized decreases were countered by increases in cattle tuberculosis in regions surrounding the culled areas again, probably because of changes in badger social structures. To complicate matters, while the badger culling study was under way in England, Irish authorities were conducting a similar study. In the Irish study, called the Four Areas Trial, proactive culling appeared effective in reducing tuberculosis in cattle herds. Moreover, unlike the British study, cattle tuberculosis did not rise in surrounding areas. Design, environment, and experimental differences may explain these contrasting results. Some experts recommend vaccinating animals as a means to control tuberculosis in cattle herds. Although several vaccines were tested, none proved more effective than the human vaccine BCG. No matter how well a vaccine does experimentally, using such a vaccine to treat cattle herds remains a challenge because its use should not interfere with other husbandry practices or with any specific diagnostic methods being used in local tuberculosis control or eradication programs. In the case of wildlife, methods for delivering vaccines, controlling dosage, and ensuring the safety of nontarget species are equally challenging. Locally targeted use of a vaccine to protect cattle or wildlife against tuberculosis appears a more practical strategy than widespread deployment. Wildlife Reservoirs for Infectious Diseases Make their Eradication Unlikely Diseases that are maintained in wildlife populations and transmitted to livestock present serious, perhaps insurmountable challenges to animal health authorities in developing disease Volume 3, Number 1, 2008 / Microbe Y 33

eradication programs. Efforts to control, let alone eliminate, specific infectious diseases from wildlife populations, especially those populations that are large and widespread, are extremely difficult. Meanwhile, the test-and-slaughter policy for controlling tuberculosis in livestock proves insufficient in regions where wildlife harbor this disease. In general, measures to prevent this disease from being transmitted to and from wildlife prove more efficient than do efforts to eliminate this disease from wildlife populations in which it is established. In such cases, experts typically need to tailor risk-reduction strategies and biosecurity practices that limit local interactions between livestock and wildlife. Further, in light of past experiences, careful risk analysis should be conducted before introducing wildlife to new geographic areas. In areas where tuberculosis is endemic in wildlife, agricultural practices such as allowing wildlife access to livestock feed may no longer be tolerable if infectious diseases such as tuberculosis are to be held in check. To better track wildlife reservoirs of tuberculosis, investigators could benefit from diagnostic methods that are suited for tracking multiple species, a better understanding of disease pathogenesis in unusual animal hosts, and better models for studying vaccine efficacy and disease transmission. Because resources are limited, organizations dealing with these disease challenges need to work collaboratively, a practice that is gaining momentum as evidenced by recent symposia on diseases moving between domestic animals and wildlife. Furthermore, the Wildlife Disease Association, the Society for Tropical Veterinary Medicine, and similar groups are calling for expanded efforts to integrate livestock production and natural resource management; to address wildlife, livestock, and rangeland health issues as part of environmental impact statements; and to use science-based approaches on projects involving wildlife and livestock. SUGGESTED READING Aldwell, F. E., D. L. Keen, N. A. Parlane, M. A. Skinner, G. W. de Lisle, and B. M. Buddle. 2003. Oral vaccination with Mycobacterium bovis BCG in a lipid formulation induces resistance to pulmonary tuberculosis in brushtail possums. Vaccine 22:70 76. Daszak, P., A. A. Cunningham, and A. D. Hyatt. 2000. Emerging infectious diseases of wildlife threats to biodiversity and human health. Science 287:443 449. Donnelly, C. A., R. Woodroffe, D. R. Cox, F. J. Bourne, C. L. Cheeseman, R. S. Clifton-Hadley, G. Wei, G. Gettinby, P. Gilks, H. Jenkins, W. T. Johnston, A. M. LeFevre, J. P. McInerney, and W. I. Morrison. 2006. Positive and negative effects of widespread badger culling on tuberculosis in cattle. Nature 439:843 846. Griffin, J. M., D. H. Williams, G. E. Kelly, T. A. Clegg, I. O Boyle, J. D. Collins, and S. J. More. 2005. The impact of badger removal on the control of tuberculosis in cattle herds in Ireland. Prev. Vet. Med. 67:237 266. O Brien, D. J., S. M. Schmitt, J. S. Fierke, S. A. Hogle, S. R. Winterstein, T. M. Cooley, W. E. Moritz, K. L. Diegel, S. D. Fitzgerald, D. E. Berry, and J. B. Kaneene. 2002. Epidemiology of Mycobacterium bovis in free-ranging white-tailed deer Michigan USA, 1995 2000. Prev. Vet. Med. 54:47 63. Palmer, M. V. 2007. Tuberculosis: a reemerging disease at the interface of domestic animals and wildlife, p. 195 216. In J. E. Childs, J. S. Mackenzie, and J. E. Richt (ed.), Wildlife and emerging diseases: the biology, circumstances and consequences of cross-species transmission. Springer-Verlag, Berlin. Schmitt, S. M., S. D. Fitzgerald, T. M. Cooley, C. S. Bruning-Fann, L. Sullivan, D. Berry, T. Carlson, R. B. Minnis, J. B. Payeur, and J. Sikarskie. 1997. Bovine tuberculosis in free-ranging white-tailed deer from Michigan. J. Wildlife Dis. 33:749 758. 34 Y Microbe / Volume 3, Number 1, 2008