Since the presence of anti-p53 antibody has been correlated with the mutation and accumulation of p53, the aim of this study was to detect anti-p53 antibody and understand its correlations with anti-Tof, -Rex, or -Tax antibody reactivity in HTLV-I infected people differing in their clinical status. A plasmid (pGEX-Tof) was constructed to express Tof recombinant protein (RP) in Escherichia coli. Serum samples from 50 asymptomatic carriers (ACs), 50 adult T-cell leukemia (ATL) and 50 HTLV-I-associated myelopathyltropical spastic paraparesis (HAM/TSP) patients were assayed for reactivity with different RPs by Western immunoblotting. The results showed that 2% of ACs, 4% of ATL patients and 6% of HAM/TSP patients had anti-p53 antibody. Therefore, anti-p53 antibody is not a useful serological marker for clinical management of HTLV-I infected people. Only 1 HAM/TSP patient had anti-Tof antibody whose specificity was further confirmed by antibody competition enzyme immunoassay. This study demonstrates that Tof protein is immunogenic in vivo, suggesting that it plays a role in the life cycle and pathogenesis of HTLV-I. The rate of anti-Rex antibody among HAM/TSP patients was significantly higher than that of ACs or ATL patients. In addition, 50% of ACs, 42% of ATL and 98% of HAM/TSP patients had anti-Tax antibody. McNemar's test showed that the presence of anti-p53 antibody did not have any correlation with the anti-Tax antibody in HTLV-I-infected people, while the correlation between anti-p53 and anti-Rex antibodies or anti-p53 and anti-Tof antibodies cannot be ruled out in this study.