Influence of erythromycin A on the microbial populations in aquaculture sediment microcosms

Degradation of erythromycin A was studied using two sediment samples obtained from the salmon and trout hatchery sites at Hupp Springs (HS) and Goldendale (GD), Washington, United States. The former site had been treated for 3 years with erythromycin-medicated feed prior to the experiments, and the latter site had not been treated with any antibiotic for at least 6 years. The two sediment microcosms treated with either N-[methyl-14C]erythromycin A or [1,3,5,7,9,11,13-14C]erythromycin A showed S-curves for erythromycin A mineralization with a prolonged lag time of 120 days, except for GD microcosms treated with [1,3,5,7,9,11,13-14C]erythromycin A. We proposed a simplified logistic model to interpret the mineralization curves under the assumption of the low densities of initial populations metabolizing erythromycin A. The model was helpful for knowing the biological potential for erythromycin A degradation in sediments. Although erythromycin A added to the two sediment microcosms did not significantly alter the numbers of total viable aerobic bacteria or erythromycin-resistant bacteria, it affected the bacterial composition. The influence on the bacterial composition appeared to be greater in GD microcosms without pre-exposure to antibiotics. PCR-RFLP and DNA sequence analyses of the 16S ribosomal RNA gene and the erythromycin esterase (ere) gene revealed that ereA type 2 (ereA2) was present in potentially erythromycin-degrading Pseudomonas spp. strains GD100, GD200, HS100, HS200 and HS300, isolated from erythromycin-treated and non-treated GD and HS microcosms. Erythromycin A appeared to influence the development and proliferation of strain GD200, possibly via the lateral gene transfer of ereA2.