Detection of Mycobacterium tuberculosis group organisms in human and mousejoint tissue by reverse transcriptase PCR: Prevalence in diseased synovialtissue suggests lack of specific association with rheumatoid arthritis
Ke. Kempsell et al., Detection of Mycobacterium tuberculosis group organisms in human and mousejoint tissue by reverse transcriptase PCR: Prevalence in diseased synovialtissue suggests lack of specific association with rheumatoid arthritis, INFEC IMMUN, 69(3), 2001, pp. 1821-1831
Infection with mycobacterial species, including Mycobacterium tuberculosis,
has long been implicated in the etiopathology of rheumatoid arthritis (RA)
on the basis of clinical and pathological similarities between tuberculosi
s and RA, Despite evidence of immune responses to mycobacterial antigens in
RA patient synovial fluid, cross-reactivity between these and host joint a
ntigens, and the presence of M. tuberculosis protein antigen in RA synovial
fluid, a definite causal association with RA has not been shown. Previous
studies from our laboratory using reverse transcriptase PCR (RT-PCR) of bac
terial rRNAs have shown RA synovium to be colonized by a diverse range of b
acteria, most of commensal origin. However, M. tuberculosis group organism
(MTG) RNA sequences were found in one RA patient tissue. Since this was con
sidered of sufficient interest to warrant further investigation, we devised
a M. tuberculosis-specific nested RT-PCR test which could be used for dete
ction of MTG in a mixed pool of bacterial crDNAs. This test was used to inv
estigate the distribution of MTG in RA synovial tissue and also non-PA arth
ritis and healthy control tissues and was also used to examine the tissue d
istribution of MTG in an acute and chronic model of M. tuberculosis infecti
on in the BALB/c mouse. MTG sequences were found in a high proportion of RA
patient synovial tissues but also in non-PA arthritis control tissues at l
ower frequency. This likely reflects trafficking of persistent M. bovis BCG
to inflamed joint tissue, irrespective of cause. MTC were not found in hea
lthy synovial tissue or the tissue of patients with undifferentiated arthri
tis. In both the acute and chronic models of infection in BALB/c mice, M. t
uberculosis was also found to have trafficked to joint tissues, however, no
signs of inflammation, arthritis, or pathology associated with M. tubercul
osis infection was seen. These combined results would argue against a speci
fic causal role of MTC in RA-like arthritis; however, their role as adjuvan
t in immune dysfunction in an innately susceptible host cannot be excluded.