return to

Paper for discussion on Bovine TB in the UK.

(Recommendations ) Written by Dr Ruth Watkins for the European Livestock Association and



Though I was a specialist in Clinical Virology I had a background in General Medicine and Infectious Disease and worked as a junior microbiologist at The London Hospital under Professor Williams and also did the M Sc in Clinical Microbiology there. What a strange coincidence that my interest in conservation has led me to become a native plant farmer in West Wales with Welsh Black Cattle, Brecknock Hill and Herdwick pure bred sheep and some poultry; yet ever since I have farmed, the infectious disease problems in livestock have loomed large and my interest in infection has a continued relevance.  I am in a Bovine TB area, but I am not sure that it is classifed a hot-spot.  I would think about half the keepers of cattle in my parish have been down with BovineTB in the last 6 years or so, but when I first came here 11 years ago the vet told me there was no Bovine TB in this area of Carmarthenshire.  My cattle are tested every year, using the SICCT test, the Single Intradermal Comparative Cervical Tuberculin skin test- this is the injection of tuberculin derived from M avium and M bovis at 2 different sites on the neck.


I have been surprised to learn that veterinary public health microbiology is not conducted in the same way as in human medicine and there are no trained clinical specialists nor is there peer reviewed best practice. The only veterinary laboratory in the UK doing work on TB funded by government grants is Weybridge, and in order to get funding for their work the scientists, few of whom are vets, must formulate hypotheses to competitively bid for research funding or carry out government, DEFRA, commissioned work to devise a new test such as to detect the gamma interferon response specific for M bovis infection using peptides derived from proteins absent in the BCG vaccine (Bacille Calmette Guerin) but expressed by M bovis.


The vets responsible for adminstering the skin tests, giving results of the gamma interferon tests to the farmers and taking decisions in the field on the infected herds are not trained in microbiology.  The EU and other bodies are rigid and prevent progress and suppleness in response to infection as new tests and strategies for control become possible. M bovis research follows in the footsteps of M tuberculosis research on human disease- in many ways useful but also problematic as the goals are different.


This might seem picky and a long introduction but I think it lies at the heart of difficulties with bovine TB.  The research scientists, gifted and knowledgeable as they are, have their limits and bias- there is no disinterested peer review by clinical specialists and updating of practice, and little understanding in the field, or drive to improve matters.  There is much poor practice such as in the case of Hallmark Boxster where the vets in the field, well meaning though they may be, cannot understand the concept of a false positive result and got themsleves into difficulties necessitating a loss of face and an expensive court case.



The Mycobacteriaceae family are fascinating and the knowledge about them grows apace as many species now have their genome completely sequenced.  It is worth looking them up on Wikipedia. They are complex, having a unique cell wall in the bacterial world and there are an ever increasing number of species recognized and many are found in the environment; they may give rise to infection with no clinical consequences or give rise to disease- disease for the most part with life long infection.  Even if those mycobacteria that cause tuberculosis are not free living in the environment they can survive as infectious organisms for a considerable period of time- it is apparent from molecular studies that M bovis can remain infectious for up to 6 months on pasture in cool moist Northern Europe.  However M bovis suspended as an aerosol in particles of 5 microns in diameter may not have a half life longer than 30 minutes. (the source of this information Professor Wellington and Professor Elwood respectively).


Infection with disease causing mycobacteri in cattle M bovis and occasionally M tuberculosis (I do not cover Johne's disease caused by Mycobacterium avium subspecies paratuberculosis)

It is interesting to consider why we or cattle and so on do not all become infected with M bovis or M tuberculosis much as one would with flu, chicken pox or measles where the route of infection is similar by respiring particles 5 microns in diameter deep into the lungs. One of the answers is that during infection with M bovis little infectious aerosol is generated; in humans it is only those with lung disease where a cavity communicates with the bronchi who shed the infectious organism in an aerosol on the breath in an infectious dose. To be at risk of infection prolonged contact in a closed space such as the household or the classroom is usually required- or sitting within a few rows of someone shedding it on their breathe on a long haul aeroplane flight,  or even more casually in a nightclub.  For cattle it would be in the shed, or tethered in lines overnight at a show, or a stay overnight in a market shed.  Transmission in the open air of the field on the breath from cow to cow would be less likely. 


Whilst infection may be transmitted in infected milk to the progeny, or spread by inoculation of wounds such as badger bites, both M tuberculosis and M bovis infect most commonly by inhalation into the lungs, whether it be TB in badgers, cattle or humans. Possibly in the case of environmental contamination with M bovis the means of infection is also by aerosolisation by falling rain on badger latrines on fields. (Professor Elwood put forward this concept.  It is not unique in nature as this is the means whereby a soil bacterium the cause of the disease melioidosis can infect humans in the areas where it is resident in the soil, in N Australia and SE Asia.) Unfortunately M bovis in badgers spreads widely in the body from the lungs in many instances of infection, to the kidneys and bladder and the bowel so that very large numbers of organisms are shed in the urine or faeces. This can happen in cattle, or humans too, but takes a longer time and the culling of cattle that are positive on regular skin testing, and treatment with anti-microbials of humans, means that few reach a stage of disease with heavy environmental contamination.  The type of disseminated infection in babies, or calves, where there is widespread seeding of tuberculomas throughout the body including the eye and the brain, fatal if untreated, is not generally considered very infectious to others.


The science laboratory under Prof Glyn Hewinson put forward the hypotheses that M bovis was not latent in cattle and could spread between cattle infected experimentally on the breath before gross lesions in the lung developed at an early stage a few weeks after infection.  I am not sure they still espouse these views though they are interesting hypotheses to investigate.  The experimental work with animals involves using a large dose of M bovis to infect their subjects, as well as the use of laboratory isolates rather than field strains direct from animal to animal and does not equate very closely with natural infections. 


However the accepted pattern of infection is one where the infected animal either suppresses the infection to a clinically latent inactive state, or progresses with a continuing active infection ultimately with disease- or may reactivate latent infection years after initial exposure.  The result in any active infection is gradual accumulation of maycobacterial load over time, usually many years.


After the initial infection of the lung macrophages upon exposure, the local lymph nodes are infected by infected macrophages moving from the primary site. A specific immune response develops which is different from that to viruses.  The mycobacterium can remain viable and multiply in the macrophage, indeed this is their 'home', unless gamma interferon is produced locally by a lymphocyte. The gamma interferon stimulates the phagosome containing the mycobacteria in the macrophage to produce a surge of oxidising molecules which kills the mycobacterium.  The production of gamma interferon is dependent on a cell mediated immune response by lympocytes that recognize the proteins of the specific mycobacteria, M bovis for example.  The molecules recognized are small breakdown products produced by macrophages and antigen presenting cells that have engulfed and digested some of the mycobacteria; these nonapeptides are presented on the surface of these cells to the gamma interferon producing lymphocytes and cause its release.  This cell-mediated arm of the immune system is the response that controls mycobacterial infection.  Either the mycobacteria can become latent shortly after initial infection or the infection can proceed slowly or even become quiescent after an active phase giving rise to disease- this latter was aimed for at sanatoria in the mountains etc before anti-microbials active against TB were discovered.   


Bacille Calmette-Guerin vaccine

BCG contains living attenuated bacilli derived originally from M bovis shed in milk and developed over 10 years during the 1920s by two French microbiologists.  The bacilli in the BCG vaccine cause a localised infection at the site of vaccination, or in the gut after ingestion (it was originally used on human neonates by the oral route), that is self limiting and does not persist in the vaccinated host more than a few months.  Vaccination with BCG ideally stimulates the cell-mediated response to M bovis or M tuberculosis.  Unfortunately this response can be derailed by prior exposure to environmental mycobacteria before vaccination that turn off the release of gamma interferon the successful vaccination induces.  This is why BCG works better in some environments than others and best if given to the young or newborn.  It has never been established that BCG protects against infection- indeed in the recent study on infection of badgers in the laboratory infection was not prevented (see "BCG vaccination reduces the severity and progression of TB in badgers" Chambers et al, Proceedings of the Royal Society B 2010). The use of the BCG has been to prevent widespread or disseminated infection after initial exposure and infection. The infection of the lung still occurs, and even disease in the lung and spread on the breath as an aerosol.


If the human or animal is already infected with M bovis or M tuberculosis before vaccination there is no response that alters the course of events. 


There is still hope that BCG may be shown to provide some degree of protection against infection in the first place in the field when exposure is to aerosols containing few bacteria, unlike the experiments in a laboratory.  There has not yet been definitive proof of this.  Though BCG has been used on a billion humans, mainly babies, it has not eliminated TB infection anywhere it has been used ("Vaccines" edited by Plotkin, Orenstein and Offit).  The hope that vaccination will eliminate M bovis from badgers after 4 generations of vaccine use is fanciful- a number of badger groups have said this in the same breath as quoting from the Chambers paper (see above).


Whilst I believe BCG vaccination in badgers will be shown to reduce infection from badger to cattle because there will be a reduced contamination of the environment with M bovis in badger latrines, there has as yet been no study in the field to show that this is so.  The vaccination in badgers is intramuscular in the rump, and it does not seem to cause them any distress when they are found in the badger trap after a night of eating peanuts- many hardly notice it. Oral vaccination would seem better in order to vaccinate more widely and cheaply, but there may be problems. Firstly repeated oral vaccination, a dominant badger swallowing many oral doses over a period of time and perhaps precluding others from eating a dose, may suppress the protective response inducing tolerance- the amount and timing of vaccine may be critical to its beneficial effect. The formulation will require the live bacilli in the vaccine to pass the stomach and a wax/fatty matrix to resist digestion in the stomach is being developed. Another animal may swallow the vaccine such as cattle and unless testing has been changed to preclude vaccinated but not infected cattle from culling there may be a problem.


Finally in this section on disease I would like to point out that mycobacteria manipulate the immune response.  When there is a great load of mycobacteria in an advanced active infection the cell-mediated immunity I have described above is turned off.  The immune system switches to the production of antibody in quantity to many mycobacterial proteins.  The antibody is useless at controlling infection.  Once in this stage of infection disease if not already present will occur and death will follow. The skin test and the gamma interferon test are both measures of cell-mediated immunity.  These become negative when the switch to antibody production is made.  Some cattle never develop a positive skin test during the course of their active infection and when culled, perhaps as aged dairy cattle, may be found to be riddled with disease- they are also likely to be infectious.  However an apparently well animal or human can be infectious on the breath for instance, having disease undetected of the lungs, or whilst having disease there may be no shedding on the breath, or this may be intermittent.  I should also point out that reinfections can occur.  A good long article to read, particularly looking at the diagram on page 206 is one by Rua-Domenech, Goodchild, Vordermeier, Hewinson, Christiansen and Clifton-Hadley called "Ante mortem diagnosis of tuberculosis in cattle: A review of tuberculin tests, gamma interferon assay and other ancillary diagnostic techniques." Research in Veterinary Science vol 81, 2006 pages 190 to 210.  


Differences of approach between man and animals

It is estimated that about one third of humans worldwide have latent TB infection, the majority of infections are M tuberculosis.  Active infection and disease are now treated with anti-microbial chemotherapy.  There is a move to identify latent TB infection and treat with anti-microbial therapy which prevents its recurrence, but this is not generally practised as yet.  Treatment with the appropriate multi-drug regime for a sufficient time prevents resistance, which is chromosomal and not plasmid transmitted as in E coli and gram negatives, which otherwise develops to single drug therapy within 6 months.  Treatment of active infection and disease, and contact tracing of infectious individuals, has made an enormous difference to the incidence and prevalence of TB in developed countries- drug treatment of an infectious individual sterilises the breath usually in about a week.  Many methods to diagnose infection, from testing x-rays and samples and biopsies for culture are used.  Testing for antibody is routine as about 10% of human cases are skin test negative.


The approach in animals, apart from elephants which is along human lines of diagnosis and treatment, is different.  It varies from doing nothing to culling whole herds of cattle. The testing of cattle by the skin test in repeated cycles has eliminated bovine TB in countries where there is no wildlife reservoir. In Europe both wild red deer in some areas and wild boar can be a source of infection for cattle, reintroducing bovine TB to the national herd.  Unfortunately both M bovis and M tuberculosis can infect a wide range of animals as well as humans. As well as elephants in contact with humans, other animals on wildlife reserves in Africa for intance and occasionally cattle herds can acquire M tuberculosis from humans, though these outbreaks are usually self-limiting.   M tuberculosis may be better adapted to humans and M bovis to animals.  Badgers are the most important reservoir of M bovis infection in wildlife in England, Wales and Ireland.  Clearly there is a significant rate of infection passing from badgers to cattle as the Krebs trials to cull some 80% of badgers reduced the herd infection rate by 16%.  This effect persisted for two years after the trapping and culling of badgers was stopped- the increase in cattle infection around the periphery of culling zones disappeared when culling ceased (see "The effects of annual widespread badger culls on cattle tuberculosis following the cessation of culling." Jenkins et al JID 2008)


The recommendations I would like to make


1.         The removal of all infected cattle unfortunately should continue at present in the effort to contain and eventually to eliminate bovine TB. 


The tests done at present do not distinguish between latently infected and actively infected cattle. 

I can foresee a time when new vaccines against TB may ensure latency even if vaccination does not protect against infection (I wonder whether it will ever be possible to protect against  infection)

Tests will be developed to detect active infection- I would expect these to detect responses to, or detect the presence of, certain M bovis encoded proteins that are excreted by the actively growing and dividing M bovis mycobacteria as distinguished from latent M bovis mycobacteria.


2.         The testing for cattle infection should be brought up to a standard of accuracy and correctness  that exists for instance in the diagnosis of BTV8 or Schmallenberg virus infection. 


At present the testing to identify and remove infected cattle is incomplete as there are no tests developed for antibody to M bovis in cattle.


The ascertainment of which cattle are infected in a herd in which a bovine TB breakdown is identified  either by a bovine with a standard skin test positive or the discovery of lesions at abattoir post mortem with culture of M bovis, is inaccurate.  In the case of a herd breakdown the severe reading of the SICCT skin test is applied, also on occasion whole herd testing using the BoviGam test (standard skin test positve cattle that have already been identified have been removed and killed before the blood is taken for the gamma interferon test).  It is inaccurate because though the standard skin test has been verified as highly specific though not highly sensitve, the severe reading has not been verified (see above the paper entitled "Antemortem diagnosis of tuberculosis in cattle: A review of tuberculin tests, gamma interferon tests and other ancillary techniques").  The BoviGam test for gamma interferon throws up many false positives.


The philosophy seems to be one test result no matter how dubious or weakly postive is enough to condemn a bovine to death.  It is never checked out (inconclusive skin tests may be repeated once 60 days later, if repeated before 60 days there may be a false negative result because of tolerance to tuberculin).  Neither repetition nor a different test is applied to verify the result.


The overall result is the removal of infected and uninfected cattle whilst leaving any cattle that are infected but not producing any reaction to the skin test, nor to the gamma interferon test if that test is used, because they have switched to antibody production and switched off their cell-mediated immunity to M bovis.    



3.         A test for antibody to M bovis in cattle should be developed. 

This would be checked by ensuring a number of different M bovis proteins were identified to confirm specificity of response (as used for elephants), a microarray for example.


4.         The gamma interferon test developed to detect the response to peptides such as ESAT-6 and CFP-10 should be used in place the BoviGam test, or to confirm any positives thrown up by the BoviGam test.

This is very much more specific and in keeping with modern testing that uses peptides to reduce false positives by crossreaction with protein soups (tuberculin or PPD, Purified Protein Derivative, is a protein soup consisting of some 200 proteins from M bovis and is thus bound to throw up false positives particularly in in-vitro tests)  This gamma interferon test is based on that developed for humans, and has been developed at Weybridge as a DIVA test when BCG vaccination of cattle is introduced.

            The gamma interferon test may be more difficult to do to a high standard, may require setting up in triplicate, taking further specimens if the controls do not work. There is no clear separation of negative from positive samples in the test.  Using tuberculin as in the BoviGam test throws up false positives from natural killer cell activity, white blood cells responsible for non-specific immune response to invading pathogens.  These can be strongly false positive.


5.         Verification of the severe reading for the SICCT skin test should be made: the severe reading should be abandoned as inaccurate if it cannot be verified

(I don't believe it is possible to measure 2mm differences in skin tests on cattle with accuracy) and the peptide gamma interferon test applied and the antibody test on infected herds.



Action in hot-spot areas


I believe culling of badgers to reduce the population (not extermination) is necessary. Incidentally it would contribute to the survival of the hedgehog which is in danger of becoming extinct in the UK.   Hedgehogs are defenceless against badgers and as they need to use the same areas at night they are found, killed and eaten completely except for their pelt of bristles licked clean- their rustling as they walk and their smell give them away.


Trapping and vaccination of badgers on the periphery of the culling zones or on the land of those that refuse to allow the cull of badgers should be added to the strategy.


This is all very expensive.


In the Intensive Action Area in Wales in North Pembrokeshire, all the measures taken by farmers and the extra restrictions and testing have reduced herd breakdowns over several years by 37%.  Culling of badgers has not yet been permitted nor badger vaccination.



Vaccination of badgers


The amount of M bovis shed into the environment in urine or faeces or sputa is less overall from badgers infected after vaccination with BCG than without vaccination.  Vaccination of badgers can be expected to reduce the number of herd breakdowns with bovine TB, but it is not yet scientifically proven to do so.  Would it be more effective than culling 70 - 80% of the badger population or be additive to the effect of culling?  One scientific field trial is underway.  The whole process of developing and using BCG vaccine for badgers has been inexcusably slow and decades late.



Vaccination of cattle


I belive that the use of the BCG vaccine will achieve very little in cattle. 


It has been shown that vaccination of calves within hours of birth by injection is as effective on immunology testing, or more effective, than vaccination at an older age.


But as it will not prevent infection by the aerosol route how many cattle will it save from culling if the policy remains that all infected cattle must be detected and killed?  If BCG vaccination does not prevent lung disease with M bovis and shedding on the breath will it reduce the number of cattle infected in a herd?


Again it is essential to do vaccine trials in the field on dairy and suckler cattle herds that are subject to natural infection.  I hope my opinion would be proved wrong.  Neither cattle nor badgers live as long as humans and if BCG vaccination, or some manipulation of it, produces some protection against infection when exposure is low, to very small numbers of infectious bacilli, then vaccination may make a difference to the numbers of cattle or badgers infected.