Advancing Desulfurization in the Model Biocatalyst Rhodococcus qingshengii IGTS8 via an In Locus Combinatorial Approach

Abstract

Biodesulfurization poses as an ideal replacement to the high cost hydrodesulfurization of the recalcitrant heterocyclic sulfur compounds, such as dibenzothiophene (DBT) and its derivatives. The increasingly stringent limits on fuel sulfur content intensify the need for improved desulfurization biocatalysts, without sacrificing the calorific value of the fuel. Selective sulfur removal in a wide range of biodesulfurization strains, as well as in the model biocatalyst Rhodococcus qingshengii IGTS8, occurs via the 4S metabolic pathway that involves the dszABC operon, which encodes enzymes that catalyze the generation of 2-hydroxybiphenyl and sulfite from DBT. Here, using a homologous recombination process, we generate two recombinant IGTS8 biocatalysts, harboring native or rearranged, nonrepressible desulfurization operons, within the native dsz locus. The alleviation of sulfate-, methionine-, and cysteine-mediated dsz repression is achieved through the exchange of the native promoter P_dsz, with the nonrepressible P_kap1 promoter. The Dsz-mediated desulfurization from DBT was monitored at three growth phases, through HPLC analysis of end product levels. Notably, an 86-fold enhancement of desulfurization activity was documented in the presence of selected repressive sulfur sources for the recombinant biocatalyst harboring a combination of three targeted genetic modifications, namely, a dsz operon rearrangement, a native promoter exchange, and a dszA-dszB overlap removal. In addition, transcript level comparison highlighted the diverse effects of our genetic engineering approaches on dsz mRNA ratios and revealed a gene-specific differential increase in mRNA levels. IMPORTANCE Rhodococcus is perhaps the most promising biodesulfurization genus and is able to withstand the harsh process conditions of a biphasic biodesulfurization process. In the present work, we constructed an advanced biocatalyst harboring a combination of three genetic modifications, namely, an operon rearrangement, a promoter exchange, and a gene overlap removal. Our homologous recombination approach generated stable biocatalysts that do not require antibiotic addition, while harboring nonrepressible desulfurization operons that present very high biodesulfurization activities and are produced in simple and low-cost media. In addition, transcript level quantification validated the effects of our genetic engineering approaches on recombinant strains' dsz mRNA ratios and revealed a gene-specific differential increase in mRNA levels. Based on these findings, the present work can pave the way for further strain and process optimization studies that could eventually lead to an economically viable biodesulfurization process.

Keywords: biocatalysis; biodesulfurization; dibenzothiophene; dsz; genetic engineering; metabolic engineering.

FIG 1

4S pathway in Rhodococcus strain IGTS8. A diagram depicting the 150-kb plasmid (pSOX) of wild-type R. qingshengii IGTS8, the dsz mRNA transcript (dszA-dszB-dszC), and the enzymatic background for biodesulfurization of dibenzothiophene (DBT) is shown.

FIG 2

Organization of dsz loci. A diagram depicting the dsz locus located within the 150-kb plasmid, pSOX, of wild-type R. qingshengii IGTS8 (P_dsz-dszA-dszB-dszC) compared to the same locus of genetically engineered isogenic strains P_kap1-dszBCA and P_kap1-dszABC is shown. Notice the gene overlap between dszA and dszB, which is present only in the native operons.

FIG 3

Growth and biodesulfurization in the presence of DBT. The growth (biomass, g/L) and desulfurization capability for wt and recombinant strains grown in the presence of 0.1 mM DBT as the sole sulfur source are compared. The calculated growth kinetic parameters are presented in the top left inset (gray = wt, red = P_kap1-dszBCA, and blue = P_kap1-dszABC). The concentrations (μM) of produced 2-HBP and DBTO₂ are quantified at four time points (24, 48, 72, and 96 h). Asterisks indicate P values (*, P < 0.05; ***, P < 0.001; ****, P < 0.0001; n = 5) for comparison of total concentrations (in μM) of 2-HBP and DBTO₂ produced by each strain.

FIG 4

Desulfurization of strains grown in rich culture media. The biomass (g/L) and desulfurization capabilities (units of DBT converted to 2-HBP and DBTO₂/mg DCW) for genetically modified P_kap1-dszBCA (red) and P_kap1-dszABC (blue) strains, using either LB or NB as the culture media, are compared to the wt strain (gray). Measurement of desulfurization activity was performed after 6, 12, and 18 h of growth.

FIG 5

Effects of different sulfur sources on growth and biodesulfurization activity. Growth curves (biomass, g/L) and desulfurization capabilities (units of DBT converted to 2-HBP and DBTO₂/mg DCW) of wt, P_kap1-dszBCA, and P_kap1-dszABC isogenic strains grown in the presence of 0.5 mM DMSO, sulfate, methionine, or cysteine as the sole sulfur sources are presented. Ethanol was used as the sole carbon source (165 mM). Measurement of desulfurization activity was performed after 20, 45, and 65 h of growth. See also Table 1.

FIG 6

Effects of different carbon sources on growth and biodesulfurization activity. The biomass (g/L) and desulfurization capabilities (units of DBT converted to 2-HBP and DBTO₂/mg DCW) for genetically modified P_kap1-dszBCA (red) and P_kap1-dszABC (blue) strains, using either ethanol, glycerol, or glucose as the sole carbon source (330 mM carbon), are compared. Sulfate was used as the sole carbon source (0.5 mM). Measurement of desulfurization activity was performed after 20, 45, and 65 h of growth.

FIG 7

Optimization of ethanol concentration. The biomass (g/L) and desulfurization capabilities (units of DBT converted to 2-HBP and DBTO₂/mg DCW) for the genetically modified P_kap1-dszBCA (A and C) and P_kap1-dszABC strains (B and D), using 2, 4.5, and 7.6 g/L ethanol as the sole carbon source, are compared. DMSO (A and B) or sulfate (C and D) were used as the sole sulfur source (0.5 mM). Measurement of desulfurization activity was performed after 20, 45, and 65 h of growth.

FIG 8

Engineered loci exhibit differential dsz gene expression. (A and B) Within-gene comparison of dszA, dszB, and dszC transcriptional levels for the wt strain (P_dsz-dszABC) and genetically engineered strains (P_kap1-dszBCA and P_kap1-dszABC) in the presence of DMSO (A) or sulfate (B) as the sole sulfur sources. The fold change is expressed relative to the expression of the wt strain (the expression of the wt strain is set to 1 AU). A logarithmic scale was used. Values represent means ± the SEM. (C and D) Within-sample comparison of mRNA expression for wt and recombinant strains. Fold changes are presented relative to the lowest gene expression of each sample (set as 1 AU; values are displayed inside the circles). For an alternative graphical representation and statistical significance, see also Fig. S3 in the supplemental material. Strains were grown in the presence of DMSO (C) or sulfate (D) as the sole sulfur source. All sulfur sources were supplemented at a concentration of 0.5 mM. Ethanol (165 mM) was used as the sole carbon source. All values were normalized to the housekeeping gene, gyrB. Number of biological/technical replicates = 2/2.