Establishing in vitro and in vivo Co-culture Models of Staphylococcus epidermidis and Enterococcus faecalis to Evaluate the Effect of Topical Fluoroquinolone on Ocular Microbes

Purpose: To establish in vitro and in vivo ocular co-culture models of Staphylococcus epidermidis and Enterococcus faecalis and to study how various concentrations of moxifloxacin affect the survival of these two endophthalmitis-causing bacteria. Methods: Standard strains of S. epidermidis and E. faecalis were used. Color detection agar plates were employed to distinguish their colonies. To establish the in vitro and in vivo co-culture models, S. epidermidis and E. faecalis were co-cultivated at different ratios for various periods. For the in vivo model, various volumes and concentrations of either a mono-culture or co-culture were inoculated into the lower conjunctival sac of rabbits. Finally, the newly developed in vitro and in vivo co-culture models were subjected to the moxifloxacin treatment to access its effect on S. epidermidis and E. faecalis. Results: When S. epidermidis and E. faecalis were cultured separately in tryptic soy broth, their growth peaked and plateaued at approximately 16 and 6 h, respectively. When they were co-cultured, the growth peak of S. epidermidis got delayed, whereas the growth peak of E. faecalis did not change. The number of E. faecalis was significantly higher in the co-culture than that in the mono-culture. Treatment with moxifloxacin in the in vitro co-culture model rapidly decreased the number of S. epidermidis cells at doses ≥ 0.125 μg/ml. In contrast, the number of E. faecalis did not change significantly up to 16 μg/ml moxifloxacin. In in vivo co-culture (at 1:1), the S. epidermidis count decreased in a pattern similar to that seen in in vivo mono-culture and was barely detectable at 24 h after inoculation. In contrast, the of E. faecalis count increased up to 16 h and then decreased. When moxifloxacin was applied (zero, one, or two times) to this model, the S. epidermidis count decreased in proportion to the number of treatments. In contrast, the E. faecalis count increased with moxifloxacin treatment. Conclusions: The in vitro and in vivo co-culture models of S. epidermidis and E. faecalis were established to determine the influence of moxifloxacin eye drops on these bacteria. The results clearly show that the moxifloxacin eye drops can make E. faecalis dominant on the ocular surface.