A. Tanaka et al., Mucosal immunity and primary biliary cirrhosis: Presence of antimitochondrial antibodies in urine, HEPATOLOGY, 32(5), 2000, pp. 910-915
We have shown that IgA-class antimitochondrial autoantibodies (AMA) can be
detected in the bile and saliva of patients with PBC, suggesting that AMA a
re secreted into the luminal fluid across bile ducts and salivary glands, T
hese data prompted us to determine whether AMA of the IgA isotype may be tr
ansported across other epithelial mucosa, Therefore, we tested for the pres
ence of AMA in the urine specimens of 83 patients with PBC and 58 non-PBC c
ontrols including healthy individuals and patients with other liver disease
s. Patients enrolled in this study had no history of renal disease, and we
confirmed there was less than 50 mug/mL of protein in each of the urine spe
cimens. Interestingly, we found that AMA were present in the urine of 71/83
(86%) of all patients with PBC and in 71/78 (91%) of patients with PBC tha
t were serum AMA positive. In contrast, AMA were not detected in any of the
58 control urine specimens. Of particular interest, AMA of the IgA isotype
was present in 57/83 (69%) of patients with PBC, and in 52 of these 57, we
found secretory-type IgA, In a nested random subgroup of urine samples, th
e prevalence of the IgA2 AMA was 6/18 (33%), significantly lower than in ma
tched serum samples, 13/16 (81%, P =.007). These data show that AMA of the
IgA isotype is secreted into urine from the uroepithelium of patients with
PBC, and support the thesis that PBC originated from either a mucosal chall
enge or a loss of mucosal tolerance.