Analysis of Dibenzothiophene Desulfurization in a Recombinant Pseudomonas putida Strain

Biodesulfurization was monitored in a recombinant Pseudomonasputida CECT5279 strain. DszB desulfinase activity reached asharp maximum at the early exponential phase, but it rapidlydecreased at later growth phases. A model two-step resting-cellprocess combining sequentially P. putida cells from the lateand early exponential growth phases was designed to significantlyincrease biodesulfurization.

INTRODUCTION

Top

INTRODUCTION

Biodesulfurization (BDS), based on the application of microorganismsthat selectively remove sulfur atoms from organosulfur compounds,appears to be a viable technology to complement the traditionalhydrodesulfurization of fuels (9, 12, 18, 21, 22). Dibenzothiophene(DBT) has been widely employed as the model compound of polycyclicorganic sulfur components in fuels. DBT desulfurization in Rhodococcuserythropolis IGTS8, a model bacterium used in BDS, is catalyzedthrough the 4S pathway composed of two-component, flavin-dependentmonooxygenases, i.e., the DszA and DszC oxygenase componentsand the DszD flavin reductase, and a desulfinase, DszB. DszCcatalyzes DBT transformation into dibenzothiophene sulfoxide(DBTO) and dibenzothiophene-sulfone (DBTO₂), whereas DszA furtheroxygenates DBTO₂ to 2-hydroxybiphenyl-2-sulfinic acid (HBPS).In the last step of this pathway, DszB hydrolyzes HBPS into2-hydroxybiphenyl (HBP) and sulfite (4, 5, 10-13, 20, 21). The4S pathway has been shown to exist in a wide variety of bacterialgenera, such as Paenibacillus, Pseudomonas, Corynebacterium,and Mycobacterium, among others (9, 11, 12, 21, 22).

Several bottlenecks that limit the commercialization of BDShave been identified, and different strategies based on designingrecombinant Dsz pathways and microorganisms have been accomplishedto alleviate such limitations (3, 7, 9, 11, 12, 14, 15, 19,22). In an attempt to expand our understanding of the BDS processat a molecular level, the activity of the three 4S route enzymesin the recombinant biocatalyst Pseudomonas putida CECT5279,which harbors the R. erythropolis IGTS8 dszABC genes in plasmidpESOX3 and the hpaC gene encoding the Escherichia coli flavinmononucleotide:NADH HpaC reductase (a DszD equivalent) insertedin its chromosome (6), has been determined.

Optimal desulfurization capacity of the P. putida recombinant cells.

Top

Optimal desulfurization capacity...

To determine the optimal BDS capacity of P. putida CECT5279,cells were cultured in a bioreactor (2 liter) using basal saltmedium, containing 2 mM MgSO₄ as a sulfur source and 20 g liter^–1glutamic acid as a carbon source, as previously described (17).Samples collected at different growth times were used for resting-celldesulfurization assays using the substrates for all the 4S routeenzymes (DBT, DBTO, DBTO₂, and HBPS). Cells (0.7 g dry cellweight [DCW]/liter) were suspended in 50 mM HEPES buffer (pH8.0) and 10 µM substrate, and desulfurization was carriedout at 30°C for 2 h. Samples (0.25 ml) were collected, andthe concentrations of DBT, DBTO, DBTO₂, HBPS, and HBP were determinedby high-performance liquid chromatography as previously described(16, 17). The BDS capability (X_BDS) of the cells was determinedas the percentage of desulfurization according to the equationX_BDS = (C_HBP,2/C_DBT,0) x 100, where C_DBT,0 is the initial concentrationof DBT (µM) and C_HBP,2 is the HBP concentration (µM)after a 2-h resting-cell assay. Figure 1A shows that the maximumX_BDS (85% BDS) was reached with P. putida CECT5279 culturesat the mid-exponential phase (cells grown for 10 h), in clearcontrast with the behavior observed for R. erythropolis IGTS8cells that reached the maximum X_BDS at the stationary phaseof growth (4). To gain some insights into this peculiar BDSbehavior, the activities of the three enzymes DszA, DszB, andDszC were monitored by resting-cell assays (samples were takenevery 5 min for half an hour) along the growth curve of P. putidaCECT5279. It has been previously pointed out that the membranetransport of all the 4S route intermediates, the intracellularconcentration of reducing cofactors, and the activity of therecombinant HpaC reductase (a DszD functional equivalent) didnot influence the desulfurization process rate (2). The activityof each enzyme (a_j) (µmol/min/g DCW) was defined as thespecific time-zero reaction rate according to the equation a_j= dC_j/(C_x x dt)_{t = 0}, where C_j is the concentration of the enzymesubstrate j (µmol/liter) and C_x is the concentration ofthe biomass (g DCW/liter) employed in resting-cell assays (8).The experimental data (C_j versus time) were fitted to a first-orderdecay exponential equation. Figure 1B shows that the activitiesof the three enzymes followed clearly different patterns alongthe growth curve, which explains why X_BDS of the recombinantcells reaches the maximum level after 10 h of growth. Thus,although after 5 h of growth the activities of DszB and DszCare near their maximum, the activity of DszA is still very lowto support an efficient desulfurization and, subsequently, isacting as the BDS limiting step. Five hours later, the cellsreach the optimal combination between the decreasing activityof DszB and the increasing activity of DszA, and therefore theyare able to desulfurize DBT at the maximum rate. After that,the activity of DszB decreases, acting as the new limiting BDSstep (Fig. 1B). Since the genes dszABC are expressed as an operonunder the control of the Ptac promoter in plasmid pESOX3 (7),the differential activity profiles shown by the DszA, DszB,and DszC enzymes along the growth curve in P. putida CECT5279can be due to a still-unknown posttranscriptional regulationmechanism, e.g., mRNA processing, translation efficiency, etc.,that deserves further studies.

View larger version (35K):
[in this window]
[in a new window]

FIG. 1. Desulfurization ability (A) and DszA, DszB, and DszC enzyme activities (B) in P. putida CECT5279 cells along the growth curve. The standard deviations of three different experiments are shown.

To study further the drastic decrease in DszB activity after5 h of growth (Fig. 1B), we checked whether it could be dueto a rapid proteolysis of DszB along the growth curve. However,the amount of DszB protein increased along the growth curve(Fig. 2), strongly suggesting that the enzyme does not disappearfrom the host cells after 5 h of growth. Thus, the inactivationof the desulfinase activity rather than the degradation of theDszB protein appears to account for the observed drop and lackof HBPS desulfinase activity after the early exponential growthphase of the recombinant P. putida CECT5279 cells. Differentmechanisms for DszB inactivation, e.g., oxidative stress (2),can be foreseen, and further work needs to be accomplished toidentify the actual reason for such inactivation.

View larger version (29K):
[in this window]
[in a new window]

FIG. 2. DszB expression pattern in P. putida CECT5279. (A) DszB production and activity along the growth curve. The bars indicate the standard deviations of three different experiments. (B) Western blotting of crude extracts from cells collected at different growth times was performed by using an anti-DszB serum. The specific proteins were identified using anti-rabbit horseradish peroxidase-labeled secondary antibody with ECL Western blotting chemoluminescent reagent (Amersham). The location of the DszB protein band is indicated with an arrow.

Designing a novel two-step desulfurization process.

Top

Designing a novel two-step...

A two-step resting-cell process was designed to increase theBDS efficiency of the P. putida CECT5279 recombinant cells (Fig.3). In the first step, P. putida CECT5279 cells (0.7 g DCW/liter)grown for 10 h, showing the maximum activities of DszA and DszCmonooxygenases, were able to transform all DBT into a 50% mixtureof HBPS and HBP after 75 min, and this transformation rate didnot increase further even after 150 min of reaction. In thesecond step, the cells were removed from the reaction medium,and 0.7 g DCW/liter of cells from a P. putida CECT5279 culturegrown for 5 h and, hence, presenting the maximum DszB activitywere added. Immediately after the addition, the remaining HBPSwas efficiently converted into HBP showing a complete transformationafter 400 min of the whole resting-cell process (Fig. 3). Itshould be noted that the second step proceeds more slowly thanthe first step, likely because the activity of DszB is partiallyinhibited by the high concentration of HBP (1).

View larger version (29K):
[in this window]
[in a new window]

FIG. 3. A two-step BDS process employing P. putida CECT 5279 cells collected at two different growth stages. The concentration of the 4S route compounds was monitored throughout the resting-cell process. The bars indicate the standard deviations of three different experiments.

In summary, we have shown in this work that the availabilityof an active DszB enzyme is a key factor that finally controlsthe optimal BDS process in a recombinant P. putida strain. Althoughthe DszB protein is produced along the growth curve at a constantrate, its activity reaches a sharp maximum at the early exponentialphase and rapidly decreases at the middle exponential phase,hampering an efficient transformation of HBPS into HBP. To overcomethis problem, we have designed a model two-step BDS resting-cellprocess using P. putida cells from the late and early exponentialgrowth phases for the first and second steps, respectively.In this way, the efficiency of the BDS process could be increasedso that the desulfurization time and biomass were minimized.This work contributes, therefore, new insights on recombinantbiocatalyst behavior and provides a glimpse of new perspectivesin BDS.

ACKNOWLEDGMENTS

This work was supported by MCyT-Plan Nacional de I+D, undercontracts CTQ2004-6553-C02-01, CTQ2007-60918/PPQ, CSD2007-00005,and GEN2006-27750-C5-3-E. The grant awarded to one of the authors(J.C.) by the Ministerio de Educación y Ciencia, andcofunded by the European Social Fund, is gratefully recognized.

FOOTNOTES

* Corresponding author. Mailing address: Departamento de Ingeniería Química, Facultad de Ciencias Químicas, Universidad Complutense de Madrid, Ciudad Universitaria s/n, 28040 Madrid, Spain. Phone: 34-913944179. Fax: 34-913944171. E-mail: vesantos{at}quim.ucm.es

Published ahead of print on 1 December 2008.

REFERENCES

Top