Removal of radioactive cobalt at trace levels (approximate to nM) in the presence of large excess (106-fold) of corrosion product ions of complexed Fe, Cr, and Ni in spent chemical decontamination formulations (simulated effluent) of nuclear reactors is currently done by using synthetic organic ion exchangers. A large volume of solid waste is generated due to the nonspecific nature of ion sorption. Our earlier work using various fungi and bacteria, with the aim of nuclear waste volume reduction, realized up to 30% of Co removal with specific capacities calculated up to 1 mu g/g in 6-24 h. In the present study using engineered Escherichia coli expressing NiCoT genes from Rhodopseudomonas palustris CGA009 (RP) and Novosphingobium aromaticivorans F-199 (NA), we report a significant increase in the specific capacity for Co removal (12 mu g/g) in 1-h exposure to simulated effluent. About 85% of Co removal was achieved in a two-cycle treatment with the cloned bacteria. Expression of NiCoT genes in the E. coli knockout mutant of NiCoT efflux gene (rcnA) was more efficient as compared to expression in wild-type E. coli MC4100, JM109 and BL21 (DE3) hosts. The viability of the E. coli strains in the formulation as well as at different doses of gamma rays exposure and the effect of gamma dose on their cobalt removal capacity are determined. The potential application scheme of the above process of bioremediation of cobalt from nuclear power reactor chemical decontamination effluents is discussed.