The objectives of this study were to determine (i) whether Chlamydia (C) abortus would adhere to the intact zona pellucida (ZP-intact) of early in vitro produced bovine embryos; (ii) whether the bacteria would adhere to the embryos (ZP-free) after in vitro infection; and (iii) the efficacy of the International Embryo Transfer Society (LETS) washing protocol. The experimentation was made twice. For each replicate 100 (8-16-cell) bovine embryos produced in vitro were randomly divided into 10 batches. Height batches (4 ZP-intact and 4 ZP-free) of 10 embryos were incubated in a medium containing 4 x 10(7) Chlamydia/ml of AB7 strain. After incubation for 18 h at 37 degrees C in an atmosphere of 5% CO2, the embryos were washed in accordance with the LETS guidelines. In parallel, two batches (1 ZP-intact and 1 ZP-free) of 10 embryos were subjected to similar procedures but without exposure to C abortus as a control group. The 10 washing fluids from each batch were collected and centrifuged for 1 h at 13,000 xg. Each batch of washed embryos and each wash pellets were tested using PCR. C abortus DNA was found in all ZP-intact and ZP-free batches of 10 embryos after 10 successive washes. For ZP-intact infected embryos, Chlamydia-DNA was also detected in all 10 wash baths for two batches (2/8) of embryos, whereas for ZP-free infected embryos, Chlamydia-DNA was detected in all 10 wash baths for 6/8 batches of embryos. In contrast, none of the embryos or their washing fluids in the control batches was DNA positive. The bacterial load for batches of 10 embryos after the 10 wash baths was significantly higher for batches of ZP-free embryos (20.7 +/- 9 x 10(3) bacteria/mL) than for batches of ZP-intact embryos (0.47 +/- 0.19 x 10(3) bacteria/mL). These results demonstrate that C. abortus adheres to the ZP as well as the early embryonic cells of in vitro produced bovine embryos after in vitro infection, and that the standard washing protocol recommended by the LETS fails to remove it.