Molecular epidemiology and diagnosis of "Mycobacterium bovis" infections in African cattle

Mycobacterium bovis is the major causative agent of bovine tuberculosis (BTB) and
part of the Mycobacterium tuberculosis complex (MTBC). BTB can have an impact
on the national and international economy, affects the ecosystem via transmission to
wildlife and is of public health concern due to its zoonotic potential. Although still
present in some industrialized countries, BTB today mostly affects developing
countries lacking the resources to apply expensive test and slaughter schemes. In
Africa, the disease is present virtually on the whole continent; however, little accurate
information on its distribution and prevalence is available. Evaluations of antemortem
tests for the diagnosis of BTB in Africa are scarce but a prerequisite to
identify appropriate tools for future disease control programs. Spoligotyping and
variable number of tandem repeat (VNTR) typing of M. bovis strains isolated from
cattle in the UK has revealed the predominance of a single clonal complex of strains
with subtypes of the complex being geographically localized to specific regions within
the country. Spoligotype patterns of strains isolated from cattle from several countries
throughout the world have been reported in a number of recent studies and the
construction of databases (www.Mbovis.org, SPOLDB4) has facilitated their
comparison and helped elucidate the distribution of specific strain families.
In an attempt to gain insights into the population structure of M. bovis in Africa, we
have isolated strains of M. bovis from cattle carcasses with gross visible lesions at
abattoirs in Dakar (Senegal), Bamako (Mali), Sarh (Chad), Morogoro (Tanzania),
Algiers (Algeria) and Blida (Algeria). These mycobacteria were subjected to
spoligotyping and VNTR typing. A specific region of difference, which we have
named RDAf1, was previously found to be absent in strains from Chad and presence
of this deletion was assessed in our strain collection by PCR. In collaboration with
others, additional strains of M. bovis from other African countries were subjected to molecular typing.
At the abattoir of Sarh in Chad, 954 cattle were subjected to single intra-dermal
comparative cervical tuberculin (SICCT) testing and two recently developed
fluorescence polarization assays (FPA) prior to slaughter. Animal carcasses
underwent standard meat inspection. Gross visible lesions were extracted, analyzed
by microscopy and cultured. Cultured acid-fast bacilli (AFB) were further
characterized by molecular techniques. The different diagnostic tests were evaluated
using a sub-population of animals with either a PCR confirmed MTBC infection or no visible lesions. In addition, a Bayesian model for the evaluation of multiple diagnostic
tests in the absence of a gold standard method was developed.
In collaboration with others, we have identified a clonal complex of strains of M. bovis
present at high frequency in cattle in population samples from Chad, Cameroon,
Nigeria and Mali. This closely related group of bacteria is defined by the RDAf1
chromosomal deletion and can be identified by the absence of spacer 30 in its
spoligotype pattern. We have named this group of strains the Bovis African1 (Af1)
clonal complex. Strains of the Af1 clonal complex were not detected in population
samples from other regions in Africa or other parts of the world, suggesting that the
Af1 clonal complex is geographically localized to sub-Saharan West-Central Africa.
VNTR typing allowed to distinguish sub-populations of the Af1 clonal complex, which
were geographically localized to different countries. This was an unexpected result
suggesting that the movement of strains between countries is not common in this
area. In Mali, in addition to Af1, a second clonal group of M. bovis has been detected
and matching VNTR patterns for some of its strains and strains from France could
indicate a French origin. In Tanzania, also two clonal complexes of M. bovis were
detected by spoligotyping with one clade showing a link to strains, previously
identified in Uganda and Ethiopia and the second clade showing a link to strains
previously isolated in South Africa. M. bovis strains isolated from Algerian cattle were
closely related to strains from continental Europe and especially France. In
conclusion, our work has revealed important insights into the population structure of
M. bovis in Africa and suggests the presence of distinct clonal complexes of strains,
geographically localized to specific areas of the continent.
Our Bayesian model estimated the true BTB prevalence amongst the slaughterhouse
cattle population in Sarh, Chad to be at 8%. The Bayesian and the gold standard test
evaluation methods indicated that the ideal cut-off for positive SICCT test
interpretation should be lowered from > 4 mm (OIE standard cut-off) to > 2 mm, in
the Chadian setting. This result is of practical relevance and likely to apply to other
countries in sub-Saharan Africa. Using this cut-off, sensitivity and specificity of
SICCT was estimated at around 65% and 90%, respectively. Both FPA tests showed
a sensitivity of less than 50% but specificities of at least 90%. Our results suggested
that a substantial amount of lesions detected at the abattoir have been caused by other organisms than M. bovis.