Previous or present infection with Shiga-like toxin producing E. coli (SLTEC) was detected by an indirect neutralization assay of antibody titer. Bovine colostra and sera blocked the cytotoxic effects of Shiga-like toxin on Vero cell monolayers. SLT neutralizing antibodies were present in 84.0% (189/225) of the colostrum samples from randomly chosen cows in Bavaria, Germany. While all of the colostra with neutralizing activity reacted with SLT-I, only 14.7% neutralized both SLT-I and -II. Approximately 93.0% (37/40) of sera from heifers had SLT neutralizing activity. To quantify the neutralizing antibodies, colostra were tested in the Vero cell assay for their capability to reduce the 50% cytotoxic dose (CD50) of SLT standards, where the neutralizing units/ml (nu/ml) equal the log10 of CD50 reduction. Almost half of reactive colostra (48.7%) reduced the CD50 of the SLT-I standard by 10(4) to 10(5) (4-5 nu/ml). Higher reactivity (5-7 nu/ml) was found in 46.5% of positive colostra. The remaining colostra samples had over 7 nu/ml. To determine if the colostra were blocking receptors for SLT on Vero cells, cells were preincubated with colostra, and SLT was later added. No neutralizing activity was detected, indicating the reactivity of colostra was directed against SLT. When the colostra were subjected to ammonium sulphate precipitation and DEAE anion exchange chromatography, high levels of neutralizing activity were found in the IgG1 containing fractions. Colostrum fractions were tested for SLT-I binding antibodies in a capture ELISA, based on the binding of SLT-I to the toxin receptor analogue P1-glycoprotein. Only fractions from colostra with over 5 nu/ml were reactive in this assay, indicating the ELISA was less sensitive than the Vero cell assay. The results support the theory that SLTEC exposure of cows in Germany is more widespread than expected from epidemiological studies based on bacterial isolation. This possibly indicates a higher risk of human SLTEC infection via beef and milk products.