Microbial desulfurization of alkylated dibenzothiophenes from a hydrodesulfurized middle distillate by Rhodococcus erythropolis I-19.

Rhodococcus erythropolis I-19, containing multiple copies of key dsz genes, was used to desulfurize alkylated dibenzothiophenes (Cx-DBTs) found in a hydrodesulfurized middle-distillate petroleum (MD 1850). Initial desulfurization rates of dibenzothiophene (DBT) and MD 1850 by I-19 were 5.0 and 2.5 micromol g dry cell weight(-1) min(-1), more than 25-fold higher than that for wild-type bacteria. According to sulfur K-edge X-ray absorption near-edge structure (XANES) analysis, thiophenic compounds accounted for >95% of the total sulfur found in MD 1850, predominantly Cx-DBTs and alkylated benzothiophenes. Extensive biodesulfurization resulted in a 67% reduction of total sulfur from 1,850 to 615 ppm S. XANES analysis of the 615-ppm material gave a sulfur distribution of 75% thiophenes, 11% sulfides, 2% sulfoxides, and 12% sulfones. I-19 preferentially desulfurized DBT and C1-DBTs, followed by the more highly alkylated Cx-DBTs. Shifting zero- to first-order (first-order) desulfurization rate kinetics were observed when MD 1850 was diluted with hexadecane. Apparent saturation rate constant (K(0)) and half-saturation rate constant (K(1)) values were calculated to be 2.8 micromol g dry cell weight(-1) min(-1) and 130 ppm, respectively. However, partial biocatalytic reduction of MD 1850 sulfur concentration followed by determination of initial rates with fresh biocatalyst led to a sigmoidal kinetic behavior. A competitive-substrate model suggested that the apparent K(1) values for each group of Cx-DBTs increased with increasing alkylation. Overall desulfurization rate kinetics with I-19 were affected by the concentration and distribution of Cx-DBTs according to the number and/or lengths of alkyl groups attached to the basic ring structure.