Selective desulfurization of dibenzothiophene by newly isolated Corynebacterium sp. strain P32C1

Selective desulfurization of dibenzothiophene by newly isolated Corynebacterium sp. strain P32C1

Biochemical Engineering Journal 5 (2000) 11±16 Selective desulfurization of dibenzothiophene by newly isolated Corynebacterium sp. strain P32C1 S. Ma...

202KB Sizes 0 Downloads 22 Views

Biochemical Engineering Journal 5 (2000) 11±16

Selective desulfurization of dibenzothiophene by newly isolated Corynebacterium sp. strain P32C1 S. Maghsoudia, A. Kheirolomooma,*, M. Vossoughib, E. Tanakac, S. Katohd

a Department of Chemical Engineering, Sharif University of Technology, Tehran, Iran Biochemical & Bioenvironmental Research Center, Sharif University of Technology, Tehran, Iran c Department of Chemical Science and Engineering, Faculty of Engineering, Kobe University, Rokkodai, Nada, Kobe, Hyogo 657-8501, Japan d Graduate School of Science and Technology, Kobe University Rokkodai, Nada, Kobe, Hyogo 657-8501, Japan b

Received 10 September 1999; accepted 15 October 1999

Abstract Corynebacterium sp. strain P32C1 was selected from 30 strains isolated from soil samples of various areas in Iran on the basis of the ability to utilize dibenzothiophene (DBT) as a sole source of sulfur. During 27 h of cultivation in a jar-fermentor, DBT with an initial concentration of 0.25 mM was completely converted to 2-hydroxybip henyl (2HBP). Concentration change of 20 -hydroxybiphenyl-2sul®nate (HBPS) during cultivation showed the general characteristics of reactions in series, DBT ! HBPS ! 2HBP. With resting cells prepared in late exponential phase, DBT of 0.5 mM was completely converted to 2HBP during 30 min, and the maximum speci®c production rate of 2HBP was 37 mmol/(kg dry cells h). Fed-batch addition of DBT in the jar-fermentor resulted in growth continuing up to a high 2HBP concentration of 0.4 mM. Desulfurization ability of P32C1 strain was compared with Rhodococcus sp. IGTS8 as a standard strain, and it was thought that P32C1 strain had a higher desulfurization activity. # 2000 Elsevier Science S.A. All rights reserved. Keywords: Biodesulfurization; Dibenzothiophene; Rhodocuccus sp. IGTS8; Corynebacterium sp. P32C1

1. Introduction Sulfur-containing compounds in fossil fuels represent major problems for the petroleum and coal industries. Without treatment at the re®ning and production stages or elimination of noxious gases on combustion, the use of oil and coal fuels results in polluting or corrosive sulfurcontaining emissions and the acid rain. Governments throughout the world have recognized the problems associated with these emissions and moved to reduce them through legislations [1,2]. The conventional method for desulfurization of middle distillates like kerosene and diesel is hydrodesulfurization (HDS). Hydrodesulfurization is operated under high-pressure (150±3000 lb/in2) and high temperature (290±4558C) and uses hydrogen gas in presence of metal catalysts to reduce the sulfur in petroleum fractions to hydrogen sul®de [3]. Investment and operating costs of this process are high and, in addition, it does not work

* Corresponding author. Tel.: ‡98-21-600-5417; Fax: ‡98-21-601-2983. E-mail address: [email protected] (A. Kheirolomoom).

well on certain sulfur compounds like polycyclic sulfur compounds. Biocatalytic desulfurization (BDS) of petroleum fractions offers an attractive alternative to HDS process. For this process to be commercial, microorganisms with high activity and selectivity are required. Dibenzothiophene (DBT) was widely used as a model sulfur compound for isolation and enrichment of suitable strains. Several microorganisms were found to desulfurize DBT through a selective pathway to 2-hydroxybiphenyl (2HBP). In this pathway (4S pathway) shown in Fig. 1, carbon bonds remain intact and the heating value of fuels remains constant. Rhodococcus sp. IGTS8 (ATCC 53968) [4,5], Corynebacterium SY-1 [6], Rhodococcus erythropolis D-1 [7], a thermophilic strain of Paenibacillus [8], and Gordona strain CYKS1 [9] were observed to selectively desulfurize DBT to 2HBP. For Rhodococcus sp. IGTS8 molecular cloning and characterization of genes responsible for sulfur oxidation have been studied [10,11,12], and the enzymatic system for desulfurization pathway has been elucidated [13,14,15,16]. In this work, growth of newly isolated strain P32C1 and biodesulfurization of DBT by growing and resting cells were

1369-703X/00/$ ± see front matter # 2000 Elsevier Science S.A. All rights reserved. PII: S 1 3 6 9 - 7 0 3 X ( 9 9 ) 0 0 0 5 5 - 8


S. Maghsoudi et al. / Biochemical Engineering Journal 5 (2000) 11±16

2.2. Bacterial strains Rhodococcus erythropolis IGTS8 (ATCC 53968), which formerly was identi®ed as Rhodococcus rhodochrous was used as a standard and well-characterized strain for comparison and control. Corynebacterium sp. P32C1 was isolated from 40 soil samples from different coal mining areas and oil production ®elds in Iran. One gram soil of the samples was suspended in 10 ml of a saline solution, and after standing for 1 h, 2 ml of the supernatants was transferred to 20 ml sulfur-free Medium A [7] supplemented with 5.4 mM DBT as sulfur source and 5 g/l glucose (Table 1). During one week cultivation, single-colony isolation was repeated on the plate of the same medium containing 1.5% agar, and ®nally, isolated colonies were streaked on the plate of Tryptic Soy Agar (TSA, Difco). Selection of P32C1 strain among 30 isolated strains was based on ability to grow in the presence of DBT and production of 2HBP from DBT. 2.3. Media and growth conditions Fig. 1. Selective pathway (4S pathway) for DBT desulfurization without cleavage of carbon±carbon bonds.

studied and its characteristics were compared with those of Rhodococcus sp. IGTS8 as a standard strain. 2. Materials and methods 2.1. Chemicals DBT (Nacalai Tesque, Kyoto, Japan), Dibenzothiophene sulfone (Aldrich, Milwaukee, Wis., USA), 2HBP (Wako, Osaka, Japan), and 2,20 -dihydroxybiphenyl (Tokyo Kasei Kogyo, Tokyo, Japan) were used. 20 -Hydroxybiphenyl-2sul®nate (HBPS) in the form of sultine (dibenz[c,e][1,2]oxathiin-6-oxide) was synthesized by the technique of Hanson and Kemp [17]. All other chemicals were of analytical grade, commercially available and used without further puri®cation.

Sulfur-free Medium A [7] containing 5 g/l glucose and supplemented with 5.4 mM DBT was used for selective enrichment and isolation of new strains. For growth study of P32C1 in jar-fermentor, glucose was replaced by the same amount of glycerol in Medium A, and DBT concentration was reduced to 0.25 mM. Growth media for IGTS8 strain was BSM medium [4] containing 20 mM glycerol and DBT of 0.25 mM (Table 1). Depending on the volume of the medium used, two different DBT±ethanol solutions of 27 and 100 mM were prepared and added to sterilized medium. For subculture and small-scale experiments 500 ml ¯asks or test tubes (18  180 mm) were shaken at 308C and 150 rpm. The cell concentration was determined from the optical density at 660 nm (OD660). 2.4. Culture in jar-fermentor Batch and fed-batch culture of strains have been performed in a jar-fermentor (MDS-U50, B.E. Marubishi, Tokyo, Japan) with working volume of 1.0 l. The tempera-

Table 1 Composition of media (without carbon source) Medium A [7]

BSM medium [4]

Medium A Deionized water (l) KH2PO4 (g) K2HPO4 (g) NH4Cl (g) MgCl26H2O (g) CaCl2 (g) NaCl (g) Metal solution (ml) Vitamin mixture (ml)

Metal solution 1 0.5 4 1 0.2 0.02 0.01 10 1

FeCl24H2O (g) ZnCl2 (g) MnCl24H2O (g) Na2MoO42H2O (g) CuCl2 (g) Na2WO42H2O (g) HCl (mmol) Deoinized water (l)

Vitamin mixture 0.5 0.5 0.5 0.1 0.05 0.05 120 1

Calcium pantothenate (mg) Inositol (mg) Niacin (mg) Pyridoxine hydrochloride (mg) P-Aminobenzoic acid (mg) Cyanocobalamin (mg) Deoinized water (l)

400 200 400 400 200 0.5 1

Deoinized water (l) KH2PO4 (g) Na2HPO4 (g) NH4Cl (g) MgCl26H2O (g) CaCl22H2O (g) FeCl36H2O (g)

1 2.44 5.57 2 0.2 0.001 0.001

S. Maghsoudi et al. / Biochemical Engineering Journal 5 (2000) 11±16

ture was kept at 308C and the dissolved oxygen concentration in the culture medium was controlled to be not lower than 50% of oxygen saturation concentration by changing the aeration rate, the ratio of pure oxygen to air and/or agitation speed. Antifoam agent (KM-70, Shin-Etsu Chemical, Tokyo) was used to suppress the foaming. 2.5. Reaction with resting cells Both strains were grown in the jar-fermentor containing Medium A with 0.25 mM DBT. Cells were harvested during growth or in the late logarithmic phase by centrifugation at 10 000  g for 10 min at 48C, washed twice with 0.1 M potassium phosphate buffer (pH 7.0). The harvested cells were directly used or stored at ÿ208C. The reaction rates with resting cells were measured in test tubes or small 50 ml screw-cap glass bottles at 308C and under rotary shaking at 150 rpm. The cells were suspended in the buffer at a desired concentration of cells. After addition of DBT±ethanol solution at the concentration of 1.0 mM, degradation of DBT and formation of 2HBP were measured (see Section 2.6) and the speci®c production rate (mmol/(kg dry cells h)) calculated. 2.6. Analytical methods The cell concentration was determined from the optical density at 660 nm (OD660). For P32C1 strain, a linear relationship between OD and dry cells weight was obtained in the range from 0.08 to 0.7 (absorbance); g dry cells (g/l) ˆ 0.39 OD ÿ 0.0037. The cell concentration of IGTS8 was calculated by taking one absorbance unit at 660 nm (1cm optical length) as 0.55 g dry cells/l [18]. The concentration of DBT and other metabolites in growth culture or resting-cell reactions were analyzed by gas chromatography (GC) and high-performance liquid chromatography (HPLC). Liquid samples were acidi®ed to pH 2.0 with 1 N HCl and extracted with ethyl acetate (1.0 vol/vol). A portion of ethyl acetate layer was centrifuged, and 5 ml of the supernatant was injected to a gas chromatography (CP9001, Chrompack International B.V., Netherlands) with a ¯ame ionization detector. The GC was equipped with a stainless still column (3.2 mm  1 m) packed with Silicon OV-17, 2% Chromosorb WAW DMCS 80/100. The ¯ow rate of nitrogen carrier gas was 15 ml/min. The column temperature was programmed from 120 to 2508C, initially kept at 1208C for 5 min and then increased at a rate of 38C/min. The injection and detector temperatures were maintained at 250 and 3308C, respectively. In HPLC analysis with a LC10AD pump and a SPD-M10A variant-wavelength UV monitor (both from Shimadzu, Kyoto, Japan), 10 ml of ethyl acetate layer was applied to a Cosmosil 5C18-AR-300 column (particle size: 5 mm, 4.6  150 mm, Nacalai Tesque). The absorbance of the ef¯uent solution was continuously measured mainly at 280 nm. The mobile phase was 50% of acetonitrile with a ¯ow rate of 1 ml/min. The


retention times of 2HBP and DBT were 5 and 16 min, respectively. Dibenzothiophene sulfone was appeared at 4 min and 20 -hydroxybiphenyl-2-sul®nate (HBPS) which in acidic solution changes to sultine (dibenz[c,e][1,2]oxathiin-6-oxide) had retention time of 1.5 min. Thin-layer chromatography (TLC) was performed on 250-mm-thick, precoated silica-gel plates containing a ¯uorescence indicator (Whatman, Germany) and developed with a 50 : 50 mixture of acetone±heptane. The location of compounds on the plate were detected by use of UV light (253.6 nm). 3. Results and discussion 3.1. Identification of the isolated strain, P32C1 Taxonomical identi®cation of the isolated strain was done by the National Collection of Industrial and Marine Bacteria (NCIMB Japan). Strain P32C1 was a mucoid isolate, grampositive, nonmotile, catalase positive and oxidase negative with short rod shape. Colonies of the strain on TSA plate were yellow. Based on these data this strain was identi®ed as a Corynebacterium sp. 3.2. DBT utilization by growing cells in shaking flask Fig. 2 shows the growth of P32C1 and IGTS8 strains in ¯asks containing 150 ml of Medium A, as well as the concentrations of DBT, 2HBP and intermediate HBPS. In the same time the growth of IGTS8 in BSM medium was also studied. The growth and desulfurization characteristics of IGTS8 strain generally were the same in both the mediums. The growth rate of IGTS8 with the form of DBT±ethanol solution (doubling time of shorter than 10 h) was higher than that with DBT in solid (doubling time of longer than 30 h [19]). As shown in Fig. 2(B,C), P32C1 strain could degrade DBT to a lower concentration and produce 2HBP at a higher concentration than those of IGTS8 strain. P32C1 strain could metabolize DBT to 2HBP also through HBPS intermediate, which accumulates in measurable level, as shown in Fig. 2(D). Synthesized HBPS contained impurities and it was not possible to determine the concentration of HBPS. Thus, Fig. 2(D) showed the measured peak areas corresponding HBPS. These data con®rm that P32C1 selectively desulfurize DBT through 4S pathway. Since the production rate of HBPS, as it was shown for enzymatic reaction of IGTS8 [16], is nearly ®ve times higher than the consumption rate, this explains why HBPS accumulates in the medium. 3.3. Batch culture of P32C1 in jar-fermentor In Fig. 3(A), growth and desulfurization characteristics of P32C1 in the jar-fermentor with the same composition of medium and DBT concentration as in small-scale shaking


S. Maghsoudi et al. / Biochemical Engineering Journal 5 (2000) 11±16

Fig. 2. Growth (A), DBT degradation (B), 2HBP formation (C), and HBPS change (D), during cultivation of P32C1 and IGTS8. DBT with initial concentration of 0.25 mM was used as source of sulfur. Symbols: *, P32C1; &, IGTS8. See text for other culture conditions.

¯ask were shown. In 19 h, 0.25 mM of DBT completely degraded and after 27 h, 2HBP reached to the maximum value of 0.16 mM. The amount of 2HBP that appeared in the medium was less than consumed DBT. It was assumed that some portion of the substrate may dissolve into cell membrane or bind to inert surface [18,20]. We checked the effect of aeration on sublimation and adsorption of 2HBP during 24 h aeration to 1 l uninoculated medium. During 24 h, 12% of 0.25 mM 2HBP was lost. Although the growing cells affect this value, this can be one reason for the discrepancy between the amount of degraded DBT and produced 2HBP. The speci®c degradation rate of DBT and the speci®c production rate of 2HBP in the early growth phase were calculated as 85.0 and 16.7 mmol/(kg dry cells h), respectively. To study the activity of resting cells at different growth phases, cells were collected from fermentor in the time interval shown in Fig. 3(B), washed and resuspended in buffer to OD660 ˆ 1.0. The conversion of 0.2 mM of DBT after 4 h by these cells are shown in Fig. 3(B) as the percentage of DBT converted and 2HBP formed. This behavior showed that more active cells should be prepared from the late growth phase. 3.4. Degradation of DBT by resting cells

Fig. 3. Time course of culture of P32C1 in jar-fermentor (A). Desulfurization of DBT by the resting-cell of P32C1 from different growth phases (B). *, growth; *, DBT; &, 2HBP; ~, pH.

Formation of 2HBP from biodesulfurization of DBT with various amount of resting cells prepared from late exponential phase of P32C1 in the jar-fermentor was investigated (Fig. 4). The activity of the resting cells was shown as 2HBP

S. Maghsoudi et al. / Biochemical Engineering Journal 5 (2000) 11±16


Fig. 5. Time course of fed-batch culture of P32C1 in jar-fermentor. *, growth; *, DBT; &, 2HBP.

2HBP, although it was shown that 2HBP higher than 0.2 mM retarded growth of IGTS8 and R. erythropolis D1 strains [18,21]. 4. Conclusion

Fig. 4. 2HBP production using cells obtained from jar-fermentor (A). Specific production rate of 2HBP by different amount of cells (B). &, 5; *, 10; &, 20; *, 40; ~, 80 g dry cells/l.

formation rather DBT degradation, since accurate measurement of the DBT concentration at high cell densities was dif®cult [7,18]. Speci®c production rate of 2HBP (mmol/(kg dry cells h)) for higher cell densities were lower, probably due to mass transfer limitations, but the maximum conversions were higher. For IGTS8 strain the maximum reported rate of 2HBP production was 6.1 mmol/(kg dry cells h) [18], but with P32C1 the higher rate of 16.6 mmol/(kg dry cells h) was achieved. In another experiment, the resting cells (30 g/ l) from culture of P32C1 with 0.1 mM DBT could completely convert DBT of 0.5 mM to 2HBP during 30 min (data not shown). The speci®c production rate of 2HBP was 37 mmol/(kg dry cells h). 3.5. Fed-batch culture of P32C1 To study the activity of growing cells during cultivation and extend of 2HBP formation by P32C1 strain, DBT was added fed-batchwise after degradation of DBT, as shown by arrows (Fig. 5). With fed-batch addition of DBT during cultivation of P32C1 in the jar-fermentor containing 1 l medium with glycerol of 5 g/l and initial DBT of 0.25 mM, added DBT was degraded, repeatedly. These results showed that P32C1 strain retained the activity of DBT degradation throughout the culture period, and as well as the possibility of growth at higher concentrations of

A new isolated Corynebacterium sp. strain P32C1 showed better activity and more ef®cient desulfurization of DBT as compared with Rhodococcus sp. IGTS8. Accumulation of HBPS intermediate and detection of small amount of dibenzothiophene sulfone in culture medium con®rmed that this new strain desulfurized DBT via 4S pathway. This means that strain P32C1 could be a much more promising biocatalyst for middle distillate desulfurization as compared to other strains that use DBT as a carbon and energy source and, thus, cause a heating value loss [22,23,24]. The results from fed-batch addition of DBT that showed the possibility of passing through the barrier of 0.2 mM 2HBP during growth phase might be used for production at a high cell density of more active cells. P32C1 also has the ability to desulfurize DBT in the presence of hydrocarbon like n-hexadecane (the paper under preparation). Acknowledgements We appreciate the support of Fellowship Section of UNESCO to complete a part of this work as a collaboration between Sharif University of Technoloy in Iran and Kobe University in Japan. References [1] D.J. Monticello, Riding the fossil fuel biodesulfurization wave, CHEMTECH 28 (1998) 38±45. [2] T. Ohshiro, Y. Izumi, Microbial desulfurization of organic sulfur compounds in petroleum, Biosci. Biotechnol. Biochem. 63 (1999) 1±9.


S. Maghsoudi et al. / Biochemical Engineering Journal 5 (2000) 11±16

[3] J.G. Speight, The Desulfurization of Heavy Oils and Residua, Marcel Dekker, New York, 1981. [4] J.J. Kilbane, K. Jackowski, Biodesulfurization of water-soluble coalderived material by Rhodococcus rhodochrous IGTS8, Biotechnol. Bioeng. 40 (1992) 1107±1114. [5] J.R. Gallagher, E.S. Olson, C. Stanley, Microbial desulfurization of dibenzothiophene: a sulfur-specific pathway, FEMS Microbiol. Lett. 107 (1993) 31±36. [6] T. Omori, L. Monna, Y. Saiki, T. Kodama, Desulfurization of dibenzothiophene by Corynebacterium sp. strain SY1, Appl. Environ. Microbiol. 58 (1992) 911±915. [7] Y. Izumi, T. Ohshiro, H. Ogino, Y. Hine, M. Shimao, Selective desulfurization of dibenzothiophene by Rhodococcus erythropolis D1, Appl. Environ. Microbiol. 60 (1994) 223±226. [8] J. Konishi, Y. Ishii, T. Onaka, K. Okumura, M. Suzuki, Thermophilic carbon-sulfur-targeted biodesulfurization, Appl. Environ. Microbiol. 63 (1997) 3164±3169. [9] S.K. Rhee, J.H. Chang, Y.K. Chang, H.N. Chang, Desulfurization of dibenzothiophene and diesel oils by a newly isolated Gordona strain, CYKS1, Appl. Environ. Microbiol. 64 (1998) 2327±2331. [10] S.A. Denome, E.S. Olson, K.D. Young, Identification and cloning of genes involved in specific desulfurization of dibenzothiophene by Rhodococcus sp. strain IGTS8, Appl. Environ. Microbiol. 59 (1993) 2837±2843. [11] S.A. Denome, C. Oldfield, L.J. Nash, K.D. Young, Characterization of the desulfurization genes from Rhodococcus sp. strain IGTS8, J. Bacteriol. 176 (1994) 6707±6716. [12] C.S. Piddington, B.R. Kovavevich, J. Rambosek, Sequence and molecular characterization of a DNA region encoding the dibenzothiophene desulfurization operon of Rhodococcus sp. strain IGTS8, Appl. Environ. Microbiol. 61 (1995) 468±475. [13] T. Ohshiro, Y. Hine, Y. Izumi, Enzymatic desulfurization of dibenzothiophene by a cell-free system of Rhodococcus erythropolis D-1, FEMS Microbiol. Lett. 118 (1994) 341±344. [14] T. Ohshiro, Y. Kanabayashi, Y. Hine, Y. Izumi, Involvement of flavin


[16] [17] [18]


[20] [21] [22] [23] [24]

coenzyme in dibenzothiophene degrading enzyme system from Rhodococcus erythropolis D-1, Biosci. Biotechnol. Biochem. 59 (1995) 1349±1351. T. Ohshiro, K. Suzuki, Y. Izumi, Dibenzothiophene (DBT) degrading enzyme responsible for the first step of DBT desulfurization by Rhodococcus erythropolis D-1: purification and characterization, J. Ferment. Bioeng. 83 (1997) 233±237. K.A. Gray, O.S. Progrebinsky, G.T. Mrachko, L. Xi, D.J. Monticello, C.H. Squires, Molecular mechanisms of biocatalytic desulfurization of fossil fuels, Nature Biotechnol. 178 (1996) 6409±6418. G. Hanson, D.S. Kemp, Convenient routes to 4,400 -functionalized oterphenyls and 2,2-functionalized biphenyls, J. Org. Chem. 46 (1981) 5441±5443. H. Honda, H. Sugiyama, I. Saito, T. Kobayashi, High cell density culture of Rhodococcus rhodochrous by pH-stat feeding and dibenzothiophene degradation, J. Ferment. Bioeng. 85 (1998) 334± 338. K.J. Kayser, B.A. Bielaga-Jones, K. Jackowski, O. Odusan, J.J. Kilbane II, Utilization of organosulfur compounds by axenic and mixed culture of Rhodococcus rhodochrous IGTS8, J. General Microbiol. 139 (1993) 3123±3129. P. Wang, S. Krawiec, Kinetic analysis of desulfurization of dibenzothiophene by Rhodococcus erythropolis in batch and fedbath cultures, Appl. Environ. Microbiol. 62 (1996) 1670±1675. T. Ohshiro, K. Suzuki, Y. Izumi, Regulation of dibenzothiophene degrading enzyme activity of Rhodococcus erythropolis D-1, J. Ferment. Bioeng. 81 (1996) 121±124. D.J. Monticello, D. Bakker, W.R. Finnerty, Plasmid-mediated degradation of dibenzothiophene by Pseudomonas species, Appl. Environ. Microbiol. 94 (1985) 756±760. S. Frassinetti, L. Setti, A. Corti, P. Farrinelli, P. Montevecchi, G. Vallini, Biodegradation of dibenzothiophene by a nodulating isolate of Rhizobium meliloti, Can. J. Microbiol. 44 (1998) 289±297. A.L. Laborde, D.T. Gibson, Metabolism of dibenzothiophene by a Beijerinkia species, Appl. Environ. Microbiol. 34 (1977) 783±790.