The pgp3 protein con sisting of 264 amino acids was expressed in 9 different fragments with each varying by 66 amino acids in the form of GST fusion proteins. When these 9 fragments along with the full length pgp3 fusion proteins were reacted with the pooled positive human antiserum sample, we found that the F6 fragment lacking the N terminal read this 66 amino acids was recognized by the human antibodies as strongly as the whole pgp3 protein was. However, no other fragments were significantly recognized although there was a minimal reactivity of F2 3 with the human antibodies. The sera pooled from mice urogenitally infected with live chlamydial organisms also strictly recognized the full length pgp3 and fragment 6.

However, the antiserum raised by Inhibitors,Modulators,Libraries immunizing mice with pgp3 fusion protein recognized all fragments, suggesting that all fragments can be immunogenic if pre sented to host immune cells. Nevertheless, the highest reactivity of the pgp3 immunized mouse serum was still teins, we found that pgp3 was no longer recognizable by the human antibodies. Although the human anti bodies recognized both pgp3 and CPAF fusion proteins with an equivalent titer in ELISA, the same anti body sample only minimally recognized the pgp3 at 1 4000 while was still able to bind CPAF even after 1 1000,000 dilution in a Western blot assay. Clearly, linearizing fusion proteins in SDS gel dra matically reduced the human antibody binding to pgp3.

However, Inhibitors,Modulators,Libraries the linearized pgp3 full length or fragment fusion proteins were significantly recognized by the mouse antibody raised by immunizing mice with GST pgp3, suggesting that pgp3 sequences were antigenic even after linearization and the lack of rec ognition of the linearized pgp3 by the human antibody was due to lack of the appropriate antibody specificities. The observations that the linearized pgp3 and its fragment fusion proteins were not recognized by antisera produced during live chlamydial infection either in human or mice or the pgp3 specific mAbs while all these antibodies recognized pgp3 in ELISA have demonstrated that these antibodies are conforma with the full Inhibitors,Modulators,Libraries length and F6, suggesting that many antibody species produced during pgp3 fusion pro tein immunization mimicked the specificities of human or mouse antibodies produced during live infection.

Indeed, Inhibitors,Modulators,Libraries two monoclonal antibodies were selected out from the pgp3 fusion protein immunized mice and both only recognized the full length pgp3 and fragment 6. These observations have demonstrated Inhibitors,Modulators,Libraries that the C terminal three quarters of pgp3 amino acid sequence is required for maintaining a conformation that is recognizable by human and mouse anti pgp3 antibod ies. 3. Human antibody recognition of pgp3 is highly conformation necessary dependant When we attempted to use Western blot to confirm the human antibody binding to the chlamydial fusion pro tion dependent.