IN-VIVO BACTERIAL PROTEASE PRODUCTION DURING PSEUDOMONAS-AERUGINOSA CORNEAL INFECTION

Purpose. To establish if active pseudomonal proteases are present in v ivo during corneal infection with Pseudomonas aeruginosa and to determ ine if the mouse strains used in these and previous studies have the a bility to mount a nonocular antibody response to the purified protease s because antibodies to the bacterial proteases were not detected prev iously during in vivo ocular infection. Methods. At certain times afte r corneal infection with P. aeruginosa, corneas were harvested and sup ernatants from the corneal homogenates were analyzed for proteolytic a ctivity by zymography and immunoreactivity by immunoblotting. The effi ciency of the extraction procedures used in these studies was determin ed by incubating uninfected corneal homogenates with the purified prot eases. The resultant supernatants were analyzed for alkaline protease and elastase activity. Additionally, mice were immunized intraperitone ally with the purified proteases with and without adjuvant to determin e if the animals could mount a nonocular antibody response. Results. C orneas infected with P. aeruginosa demonstrated the presence of alkali ne protease, but not elastase, by the two methods examined. The kineti cs of the in vivo alkaline protease response closely parallels previou sly reported bacterial clearance studies in that peak alkaline proteas e activity was detected in corneal tissue when peak bacterial numbers also were observed in the eye, and it was absent when the eyes were st erile or nearly sterile. In addition, C57BL/6J mice were capable of mo unting a nonocular antibody response to microgram quantities of both p roteases only in the presence of adjuvant. Conclusions. In the model d escribed, enzymatically active alkaline protease, but not elastase, wa s demonstrated in corneal tissues during in vivo infection. Concentrat ions of these proteases were much lower than those required to stimula te an antibody response.