Persistence of microbiological agents on corroding biofilm in a model drinking water system following intentional contamination

Persistence of vegetative and spore forming bacteria on corroded iron surfaces in drinking water was studied using biofilm annular reactors operated under oligotrophic conditions. Reactors were inoculated with either Klebsiella pneumoniae or Bacillus globigii in spore form, and persistence was monitored in the bulk and biofilm phases. In the first phase of experimentation using K. pneumoniae, an initial cell concentration of 106 MPN/ml in the bulk water phase resulted in significantly longer adhesion than initial concentrations one and two orders of magnitude lower. K. pneumoniae cultured in low nutrient growth medium persisted longer in dechlorinated tap water than those cultured in full strength medium. Cell surface charge was more negative under low nutrient conditions, and this influenced electrostatic attraction between the cells and the oxidized iron surface. Cells grown in full strength media persisted longer in water with both low (<0.2 mg/L) and high (>0.5 mg/L) free chlorine residuals. Growth media injected with the cells dechlorinated the water allowing adhesion without inactivation. Microelectrode measurements showed a 40-70% drop in free chlorine from the bulk to the coupon surface, which decreased disinfectant potency against adhered cells. Growth and injection conditions clearly influenced cell adhesion and persistence, but permanent colonization of the corroded iron surface by K. pneumoniae was not observed.;The second phase of experimentation examined persistence of B. globigii (in spore form) on corroded iron. Spores were inoculated at 106 CFU/ml in the reactor bulk water, which was dechlorinated for the first experiment. Dechlorination allowed observation of the effect of shear and biofilm sloughing on persistence. Approximately 50% of the spores initially adhered were not detected after one month of operation. Addition of 10 mg/L of free chlorine after one month led to an immediate 2 log reduction, but B. globigii concentrations stabilized shortly thereafter. Increasing the chlorine concentration to 25 or 70 mg/L had no effect on inactivation. B. globigii was also injected in the presence of a normal chlorine residual (0.8 mg/L), which resulted in a steady reduction of B. globigii over one month, but density on the coupons stabilized at the same levels as the dechlorinated reactor after decontamination. Adding elevated chlorine levels after 1 month (10, 25 and 70 mg/L) had no effect on the coupon concentration. Starting decontamination immediately after injection still resulted in stabilization of the cell counts. This indicates that free chlorine cannot reach portions of the corroded iron surface where B. globigii spores are adhered. K. penumoniae presumably adhered to the same biofilm areas as the spores, but could not survive in the oligotrophic drinking water environment. As B. globigii spores are dormant and not respiring, they will persist indefinitely even in the presence of high levels of free chlorine when adhered to a tortuous surface like corroded iron.