Tuberculosis as a zoonotic disease

Blogging on Peer-Reviewed Research Tuberculosis in humans is most commonly caused by Mycobacterium tuberculosis, a slow-growing, waxy, rod-shaped bacterium. Transmitted primarily via the air when an infected individual coughs or sneezes, it’s estimated that a third of the world is infected with this agent, which causes approximately 2 million deaths every year. Though most infections are asymptomatic, infection is becoming increasingly deadly, due both to the spread of highly antibiotic-resistant strains and due to the increasing number of individuals with both HIV and TB.

While M. tuberculosis is primarily a human disease, like the MRSA I mentioned yesterday, it has the potential to be zoonotic as well–to move between animals and humans. This has been documented for a number of animal species, but one of the best-studied examples happens to be elephants, of all things–including several housed at zoos and circuses. More after the jump.

First, I’ll note that M. tuberculosis isn’t the only possible agent of tuberculosis. Though it’s by far the most common cause of the disease in humans, it’s much more rare in animals. In non-human species, other mycobacteria (such as Mycobacterium bovis) are more common causes of TB. M. bovis is also a zoonotic pathogen, occasionally infecting humans via contact with infected animals or their products (such as unpasteurized milk from M. bovis-infected cattle). So, it’s not unheard of for animals to contract tuberculosis; it’s just infrequently caused by the M. tuberculosis species, which makes the elephant reports more interesting (and, given the sad state of elephant conservation, all the more sad as well).

Almost 10 years ago, an Emerging Infectious Diseases paper came out detailing M. tuberculosis as a zoonotic disease transmitted between people and elephants. This wasn’t the first report of M. tuberculosis infection in elephants; PubMed has articles that go back to at least 1963 mentioning tuberculosis in elephants (though that one’s in Russian with no abstract). However, this was the first to document transmission between elephants and humans, during an outbreak in 1996:

In March 1996, five elephants from an exotic animal farm in Illinois were in California as part of a circus act. One elephant (with chronic, unexplained weight loss since October 1995) died under anesthesia on August 3, 1996, during a diagnostic dental work-up. Necropsy showed widespread consolidation of lung tissue with caseous necrosis of the lungs and mediastinal lymph nodes. Short, fat, relatively scant numbers of acid-fast bacilli were observed in necropsy tissues. A presumptive diagnosis of M. tuberculosis was made. The remaining four elephants were recalled to the farm in Illinois. A second elephant died en route on August 6, 1996. Necropsy revealed copious respiratory and trunk exudates and caseous necrosis of the lung.

To determine the risk for and possibility of infection among the animal trainers and caretakers, an epidemiologic investigation was initiated. The remaining elephants in the herd and the elephant handlers and trainers who were still traveling were recalled to the farm and examined for evidence of M. tuberculosis infection. All elephants were empirically begun on antituberculous therapy in early December 1996.

In all, 12 of the employees had positive PPD tests (indicating prior exposure to M. tuberculosis, and M. tuberculosis was cultured from the sputum of one handler. When it was compared genotypically with isolates taken from 3 elephants who had died of TB and an additional animal that was infected but still alive, it was found to be identical or similar to the elephant isolates, suggesting a transmission event between the two species.

A later paper looked at several different animals suspected of M. tuberculosis infection at the Los Angeles zoo: 3 goats, 2 elephants, and a black rhinoceros. Bacterial isolates from all animals were obtained, and employees were screened for infection. 55 of them were found to have evidence of TB infection (a positive skin test), but none of them appeared to have an active infection when chest x-rays were taken.

Regarding transmission to the zoo-going public, the authors suggest:

Given the public’s distance from the animals and the absence of active TB among employees with closer contact with these animals, M. tuberculosis was likely not transmitted from humans to animals at this zoo.

However, in a circus environment, things may be different. Many circuses offer elephant rides or chances to touch the animals, providing opportunity to be in close contact with the animal (and putting circus-goers at a risk of contracting TB from infected animals). Leslie Griffith reports that a number of these elephants were found to carry Mycobacterium tuberculosis, including antibiotic-resistant strains, and that handlers were also infected–and that these incidents were minimized by the circus owner.

As far as I could tell by going through the published literature, no one in the general public has been identified as having contracted TB via an elephant or other infected circus or zoo animal, but it’s difficult to really tell what the risks are if the owners aren’t forthcoming about an animal’s TB status. Of course, if an animal tests positive, they’re supposed to be kept away from other animals, and certainly away from the general public, but an animal may be infected for some time before the infection is recognized. Additionally, even when it’s diagnosed, the owners may continue to use the animal rather than risk losing money. From everything I’ve read, a circus elephant’s life is pretty terrible to begin with; it’s even more troubling that one side effect of that life could be transmission of an infection to the unaware circus goers.


Michalak et al.. 1998. Mycobacterium tuberculosis infection as a Zoonotic Disease: Transmission between Humans and Elephants. Emerging Infectious Diseases. 4:283-7. Link.

Oh et al. 2002. Human Exposure following Mycobacterium tuberculosis Infection of Multiple Animal Species in a Metropolitan Zoo. Emerging Infectious Diseases. 9: Link.

Image from

13 Replies to “Tuberculosis as a zoonotic disease”

  1. If the elephants were zoo, circus or other domesticated animals, they were more likely infected by human handlers before being shipped from one country to another. It is not really fair to call it “zoonosis”, when the animals are merely a middleman in a human to animal to human chain, and not the “natural host” of the M. tuberculosis species of bacteria.

    For most bacterial and viral organisms that are pathogenic to humans, we know very little about their true natural hosts and geographic ranges. We pay attention to them when they make humans sick and do not “see” them in their natural environments. Even with the strains of E. coli O157:H7 which encode shiga toxin, we often fail to note that the toxin is encoded by a prophage (virus of bacteria) and not by the E. coli itself. We know little about the natural host range of this Shiga phage, nor much about how the bacteria we call “Shigella” evolve from E. coli by becoming obligate host-specific parasites. Likewise, as far as I know, it is not yet clear whether Mycobacteria bovis evolved from Mycobacterium tuberculosis (we gave the disease to cattle sometime in our 15,000 year domestication process) or if M. tuberculosis evolved from M. bovis (the cattle gave it to us).

    Only since the 1980s have we developed the tools to begin to answer these questions. It is a very new field of science, and very interesting.

  2. Brian,

    I agree with everything you said except not calling it a zoonosis. Indeed, animals are incidental hosts, but so are humans with West Nile or Ebola, and we certainly call them zoonotic diseases. TB is rarely zoonotic, yes, but it certainly has the potential to be so, even if it’s just as a pit stop in transmission from one human to another.

  3. At the zoo where I volunteer, everyone used to be required to have an up-to-date negative PPD if they worked in any animal area. Some years back the requirement was changed to only primate and elephant areas. The primate requirement was a no brainer, but I didn’t understand the elephant requirement. Now it makes sense.

    The worry is not that the animals will infect us but that we will infect them. The chimps (and probably the other primates as well) get TB tested each year.

    BTW, I can’t work in primates if I am coming down with a cold either. Fortunately I’m not very sickly.

  4. What interested me was the picture appended by you is that of an Asian Elephant. I searched Google for tuberculosis in elephants and found that more cases have been reported from Asian that from African elephants. An organization called Elephant Care international is help study this problem in Nepal and in India. From the bibliography appended by them it would appear that tuberculosis has been reported from Indian Elephants in 1928.

    A paper from India states that “Phylogenetic analysis based on IFN-γ protein sequences revealed a closer relation of Asian elephants and nine-banded armadillo” (Sreekumar et al (2007) Veterinary Immunology and Immunopathology 118, 75-83. It would appear that the disappearance of the Mastodon may also be attributed to the disease (Rothschild and Laub (2006) Hyper disease in the late Pleistocene: validation of an early 20th century hypothesis Naturwissenschaften, 93: 557-564 )

  5. Wonderful posting and enlightening comments, especially Brian and Arun’s comments. Mastadon TB infected meat consumed by our spear throwing ancestors, that’s an interesting food chain! Is there any literature which links TB infected milk consumption with human disease? Also, mycobacterium avium can cause disease in humans, anything on it’s transmission from birds to man?

  6. one more reason to stick to human-only circuses like Cirque du Soleil…*grin*

    as to the link between TB infected milk and disease in humans, I do know of several people who test positive for TB, but have no signs of lung involvement, as a direct result of drinking raw milk in childhood, but that probably doesn’t answer your question, does it?

  7. Hi Leslie–

    I did mention you by name and link one of your stories in the post above.

    TB from milk, by the way, used to be a serious problem before pasteurization came along, and it’s still seen infrequently here in the US (more frequently in immigrants).

  8. There is also a fishborne form of mycobacterium that is potentially transferable to humans. Some tropical ornamental fish carry it, and humans can catch it through open sores. In the fish-keeping community, it is known as Fish-TB, but I do not know what the official name of either the human or piscine disease is. I do know that it is almost invariably lethal in fish and requires long-term antibiotics in humans.

  9. Fish fancier’s finger! (That’s not the official name. The official name you’re probably thinking of is M. marinum granuloma, and the sort of intermediate name is fish tank granuloma.) I don’t know what it’s called in fish.

    Lab Lemming, I think you’re thinking correctly of TB, but exposure to (arctic) marine mammals is also associated with M. phocacerebrale infection (“seal finger”). Maybe the reason you’re thinking of hepatitis is that eating polar bear liver can cause vitamin A toxicity?

Leave a Reply

Your email address will not be published. Required fields are marked *