In 1975 Hodson et al.[13] reported that pigs with mechanically obstructed urinary tracts produced autoantibodies to a kidney protein, Tamm-Horsfall glycoprotein (THP). That report stimulated research on the role of THP in the pathogenesis of tubulointerstitial nephritis, and awoke our interest in the possibility of detecting autoantibodies to THP in humans with urinary tract infection (UTI). We later reported that all humans have such autoantibodies [8], and that the analysis of anti-THP antibodies could be used as a tool in the workup of the patient with UTI. Patients with acute pyelonephritis increased their IgG and IgA anti-THP antibody levels, but not those with acute cystitis [8, 11]. Further, in an infectious-free period, patients with renal damage had significantly lower values than both healthy controls and patients with previous pyelonephritis but no renal damage [5, 7, 11].

THP has also been shown to be an excellent marker of abnormal urinary flow and extravasation [4, 22]. In experimental obstruction of the urinary tract, THP is found within hours in Bowman’s space and in the in-terstitium [22]. The possibility that THP autoimmunity could be part of the pathogenesis of renal damage has been studied in several experimental systems, where animals were injected with THP or urine. Rats injected with THP developed a moderate immune-complex nephritis, but no damage [14]. Rabbits, however, injected with either THP, or urine, not only developed