Journal of Biotechnology 127 (2007) 222–228

Short communication

Biodesulfurization of DBT in tetradecane and crude oil by a

facultative thermophilic bacterium
Mycobacterium goodii X7B
Fuli Li, Zhengzhi Zhang, Jinhui Feng, Xiaofeng Cai, Ping Xu ∗
State Key Laboratory of Microbial Technology, Shandong University, Jinan 250100, PR China

Received 22 March 2006; received in revised form 27 June 2006; accepted 3 July 2006

Abstract

Mycobacterium goodii X7B, a facultative thermophilic bacterium, cleaving the C–S bond of dibenzothiophene via a sulfur-
specific pathway, was investigated for DBT in tetradecane and crude oil desulfurization. The extent of growth was improved by
fed-batch culture controlled at a constant pH. The total sulfur level of dibenzothiophene in tetradecane, was reduced by 99%,
from 200 to 2 ppm within 24 h at 40 ◦ C. After 72 h treatment, 59% of the total sulfur content in Liaoning crude oil was removed,
from 3600 to 1478 ppm.
© 2006 Elsevier B.V. All rights reserved.

Keywords: Biodesulfurization; Dibenzothiophene; 2-Hydroxybiphenyl; 2-Methoxybiphenyl; Tetradecane; Crude oil

1. Introduction are known to be able to remove sulfur from DBT

Direct combustion of fossil fuels leads to sul-
fur oxide emission that contributes to acid rain and
air pollution. Dibenzothiophene (DBT), especially
its derivates bearing alky groups at the vicinity of
the S atom, are widely recognized as recalcitrant
to hydrodesulfurization (HDS) (Monticello, 2000).
A small number of genera, Arthrobacter, Bacillus,
Corynebacterium, Rhodococcus, and Mycobacterium
∗ Corresponding author. Tel.: +86 531 88564003;
fax: +86 531 88567250.
E-mail address: pingxu@sdu.edu.cn (P. Xu).
via a sulfur-specific pathway (Kilbane and Jackowski,
1992) or the extended “4SM” pathway (Li et al.,
2003).
Until now, conventional refining processes have
been performed at much higher temperature, therefore
thermophilic biodesulfurizaiton is desirable and could
be easily integrated into the refining process without
cooling the stock to 30 ◦ C (Konishi et al., 1997). More-
over, thermophilic biodesulfurization also reduces the
viscosity of crude oil, which makes the development of
crude oil biodesulfurization more practicable (Borgne
and Quintero, 2003). DBT-containing hydrocarbon was
desulfurized by the growing cells of Pseudomonas sp.

0168-1656/$ – see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jbiotec.2006.07.002
 F. Li et al. / Journal of Biotechnology 127 (2007) 222–228
(Setti et al., 1995), as well as by the growing and rest-
ing cells of Gordona sp. (Rhee et al., 1998; Chang
et al., 2000) at 30 ◦ C. Setti et al. (1992) reported an
n-alkane-degrading Pseudomonas sp. was utilized to
degrade the sulfur compounds in a deasphaltenated
heavy oil. Fedorak and Westlake (1983,1984) have
used mixed microorganisms to degrade sulfur in Prud-
hoe Bay crude oil.
This paper describes the effects of DBT and
2-hydroxybiphenyl (2-HBP) concentrations on
Mycobacterium goodii X7B cell growth as well as
the desulfurization of DBT. Biocatalysts with high
desulfurization activity were obtained from fed-batch
culture in bioreactors and applied to the desulfurization
of DBT in tetradecane and crude oil.
2. Results and discussion
2.1. Effects of DBT and 2-HBP concentrations on
the growth of strain X7B
M. goodii X7B grew better with DBT at concen-
trations lower than 0.5 mM (Fig. 1). With 0.05 mM
DBT as the sole sulfur source, 0.7 g dry cell l−1 was
Fig. 1. Effects of DBT concentration on bacterial growth (hatched
bars) and the formation of 2-HBP (filled bars) by the growing cells
of Mycobacterium goodii X7B. The bacterial cultivation was carried
out with 180 rpm at 45 ◦ C for 24 h. Data are represented as the means
of triplicate cultures ± standard error.
223
Fig. 2. Effects of 2-HBP concentration on bacterial growth (hatched
bars) and residual DBT (filled bars) by the growing cells of Mycobac-
terium goodii X7B. The bacterial cultivation was carried out with
180 rpm at 45 ◦ C for 24 h. The initial DBT concentration was 0.2 mM.
achieved within 24 h due to the insufficiency of sul-
fur source. The highest cell yield was achieved when
supplemented with 0.2 mM DBT as the sole sulfur
source, with 1.7 g dry cell l−1 after 24 h. With 0.5 mM
DBT, up to 90% DBT was degraded within 48 h. With
1.0 mM DBT, 75% DBT was eliminated within 48 h,
and 0.22 mM 2-HBP was detected. One third of 2.0 mM
DBT was degraded within 48 h (data not shown). When
the concentration of DBT was 5 mM, bacterial growth
was fully inhibited due to the toxic effect of DBT.
Therefore, 0.2 mM DBT was selected for the prepa-
ration of biocatalyst for future experiments.
The effects of 2-HBP on the growth and DBT desul-
furization by the growing cells of M. goodii X7B were
also investigated. Fig. 2 indicates that as the concentra-
tion of 2-HBP increased, the cell yield decreased due to
the toxic effect of 2-HBP. When more than 0.2 mM 2-
HBP was added into the medium, bacterial cell growth
was inhibited significantly. At the same time, the desul-
furization of DBT also decreased as the concentration
of 2-HBP increased.
2.2. Preparation of biodesulfurization catalyst
from strain X7B in fermentor
The fermentation of strain X7B was performed in
a B5 automatic bioreactor (B. Braun Co., Germany).
Cells grew well when the culture was maintained at a
constant pH with carbon and nitrate feeding (Fig. 3a
and b). We gradually increased stir speed to keep
224 F. Li et al. / Journal of Biotechnology 127 (2007) 222–228

Fig. 3. Fed-batch culture of Mycobacterium goodii X7B. (a) Dry cell weight (circle), production of 2-HBP (triangle), and desulfurization of DBT
(square); (b) the curves of parameters monitored on-line. Concentrated solution of glucose and ammonium nitrate was fed. Glucose concentration
was controlled at 5 g l−1 and the pH was kept at 7.0.

dissolved oxygen at a relatively constant value of 20%
in the fermentor for strain X7B growth, for it is known
that air supply intensely affects the results of cell yield
and the desulfurization of DBT because each incor-
porated oxygen atom to DBT is derived directly from
molecular oxygen (Oldfield et al., 1997). After 24 h fed-
batch cultivation, about 3.3 g dry cell l−1 was obtained.
Honda et al. (1998) reported the desulfurization of
DBT by R. erythropolis in fed-batch culture using a
jar-fermentor, in which 0.54 mM of DBT was used as
the sole sulfur source and acetic acid and ammonia
were fed; the final cell concentration was only 1.1 g l−1 .
And the low concentration of cells obtained may have
been due to the accumulation of 2-HBP, which inhibited
cell growth (Li et al., 1996; Ohshiro et al., 1996). In
this study, 2-HBP was not produced consistently with
the amount of DBT utilized. It was reported that 2-
HBP, the product of DBT desulfurization, was further
methylated to 2-methoxybiphenyl by M. goodii X7B
(Li et al., 2003).
 F. Li et al. / Journal of Biotechnology 127 (2007) 222–228 225

Fig. 4. Effects of cell concentration and oil water ratio on the desulfurization of DBT in tetradecane by the resting cells of Mycobacterium goodii
X7B. Reaction conditions were as follows: different cell concentrations indicated in the figure, 2% glucose, different water oil ratios indicated
in the figure, 40 ◦ C, 100 mM phosphate, and pH 7.0. Filled bars are the sterilized controls, while hatched bars are the DBT left.

2.3. Desulfurization of DBT in the presence of at 30 ◦ C, in which 5 mM DBT decreased to 2.2 mM

hydrocarbon
Tetradecane containing DBT was chosen in this
study. The effects of cell concentration and oil water
ratio on the desulfurization of DBT in tetradecane were
investigated (Fig. 4). It was found that with OD620 of
40 (12 g dry cell l−1 ) and oil water ratio of 9, most of
the 4.8 mM DBT (200 ppm S) could be removed in
3 h. When tetradecane containing 4.8 mM DBT was
brought into contact with the resting cells of X7B,
the total sulfur in tetradecane decreased from 200 to
2 ppm within 24 h, showing a 99% reduction (Fig. 5a),
and 0.17 mM sulfate was detected in the aqueous phase
(Fig. 5b). However, it was observed that 80% of DBT
was eliminated in the first 2 h reaction. Then the desul-
furization rate decreased due to the accumulation of
2-HBP and the depletion of reducing power (Ohshiro et
al., 1996). At the same time, 2-MBP was also detected
from the reaction system.
Gordona sp. CYSK1 could grow using DBT as
the sole sulfur source in the presence of hexadecane
after 48 h (Rhee et al., 1998). Using resting cells
of CYSK1, 12 mM DBT in hexadecane was desul-
furized to 8 mM in 15 h, and the specific desulfu-
rization rate was 0.066 mg sulfur g dry cell wt.−1 h−1 .
Nevertheless, in the first 3 h, the desulfurization
rate was 0.16 mg sulfur g dry cell wt.−1 h−1 (Chang
et al., 2000). In contrast, M. goodii X7B could
desulfurize DBT in the presence of tetradecane at
40 ◦ C with a much higher specific desulfurization
rate at 0.56 mg sulfur g dry cell wt.−1 h−1 for the first
2 h. This may result in a more efficient biodesul-
furization process due to the higher mass-transfer
rate achieved at 40 ◦ C (Adams and Kelly, 1998).
However, the desulfurization rate also deceased to
0.052 mg sulfur g dry cell wt.−1 h−1 for 24 h.
2.4. Biodesulfurization of crude oil
While strain X7B was shown to metabolize DBTs,
benzothiophenes, and thiophenes in diesel oil and gaso-
line desulfurization experiments (Li et al., 2003, 2005),
226 F. Li et al. / Journal of Biotechnology 127 (2007) 222–228

Fig. 5. Desulfurization of DBT in tetradecane by the resting cells of Mycobacterium goodii X7B. (a) Residual DBT (filled triangle), 2-HBP
(open square) and 2-MBP (open circle) were determined during DBT desulfurization. Control (filled square) with sterilized X7B cells was
also included; (b) production of sulfate in resting cell system (filled triangle) and sterilized control (open triangle). Reaction conditions were
as follows: 12 g dry cell wt. l−1 , 2% glucose, 9:1 water oil ratio, 40 ◦ C, 100 mM phosphate, and pH 7.0. Data are represented as the means of
triplicate experiments ± standard error.

it was of interest to determine if strain X7B could
be adopted in the biodesulfurization of crude oil. In
resting cell system, the total sulfur content of Liaon-
ing crude oil 3600 (LCO3600) decreased from 3600
to 1478 ppm after 72 h, corresponding to a reduction
of 59%. Gas chromatography with pulsed flame pho-
tometric detector (GC-PFPD, Varian Associates) was
also used to evaluate the sulfur removal from crude
oil. As shown in Fig. 6, more than half of the sulfur
compounds were removed, remaining the very high
molecular sulfur compounds in the bio-treated crude
oil.
Until now, only a few results have been reported
on crude oil desulfurization. An n-alkane-degrading
 F. Li et al. / Journal of Biotechnology 127 (2007) 222–228 227

Fig. 6. GC-PFPD chromatograms of the control and the Mycobacterium goodii X7B desulfurized crude oil. FK is the control sample without
inoculation of bacteria, and the retention time of DBT was 18 min. F7 is Mycobacterium goodii X7B desulfurized crude oil. Reaction conditions
were as follows: 12 g dry cell wt. l−1 , 2% glucose, 20:1 water oil ratio, 40 ◦ C, 100 mM phosphate, and pH 7.0.

Pseudomonas sp. strain was utilized to degrade the sul-
fur compounds in a deasphaltenated heavy oil (Setti
et al., 1992). Fedorak and Westlake (1983, 1984)
have used mixed microbial populations to degrade
sulfur heterocycles in Prudhoe Bay crude oil. The
wild type strain of Rhodococcus sp. IGTS8 was also
investigated for the biodesulfurization of crude oil,
but no decrease in total sulfur content was observed
(Kaufman et al., 1999). A facultative thermophilic bac-
terium, M. goodii X7B, could selectively desulfurize
crude oil via a sulfur-specific pathway. A significant
decrease of sulfur content in crude oil (59%) was
obtained.
It was reported that up to 87% of the Chinese oil
fields were flooded as a means to increase oil recov-
ery (Tan, 1995), especially in old Chinese oil fields
where there was more than 85% water in the recovered
crude oil (Zhou and Zhang, 2004). Therefore, it would
be more practical to incubate microorganisms directly
in this water/oil system before removing the water. In
this study, M. goodii X7B was employed in crude oil
desulfurization, and a good reduction was achieved.
We are currently investigating the enzymatic and
genetic characteristics of this moderately thermophilic
bacterium.
Acknowledgments
This work was supported by the National Natural
Science Foundation of China (Grant Nos. 20590368,
30400008).
References
Adams, M.W.W., Kelly, R.M., 1998. Finding and using hyperther-
mophilic enzymes. Trends Biotechnol. 16, 329–332.
Borgne, S.L., Quintero, R., 2003. Biotechnological processes for
the refining of petroleum. Fuel Process. Technol. 81, 155–
169.
Chang, J.H., Chang, Y.K., Cho, K.S., Chang, H.N., 2000. Desulfur-
ization of model and diesel oils by resting cells of Gordona sp.
Biotechnol. Lett. 22, 193–196.
228 F. Li et al. / Journal of Biotechnology 127 (2007) 222–228
Fedorak, P.M., Westlake, D.W.S., 1983. Microbial degradation of
organic sulfur compounds in Prudhoe Bay crude oil. Can. J.
Microbiol. 29, 291–296.
Fedorak, P.M., Westlake, D.W.S., 1984. Degradation of sulfur het-
erocycles in Prudhoe Bay crude oil by soil enrichments. Water.
Air. Soil. Pollut. 21, 225–230.
Honda, H., Sugitama, H., Saito, I., Kobayashi, T., 1998. High cell
density culture of Rhodococcus rhodochrous by pH-stat feed-
ing and dibenzothiophene degradation. J. Ferment. Bioeng. 85,
334–338.
Kaufman, E.N., Borole, A.P., Shong, R., Sides, S.L., Juengst, C.,
1999. Sulfur specificity in the bench-scale biological desulfur-
ization of crude oil by Rhodococcus IGTS8. J. Chem. Technol.
Biotechnol. 74, 1000–1004.
Kilbane, J.J., Jackowski, K., 1992. Biodesulfurization of water-
soluble coal-derived material by Rhodococcus rhodochrous
IGTS8. Biotechnol. Bioeng. 40, 1107–1114.
Konishi, J., Ishii, Y., Onaka, T., 1997. Thermophilic carbon-sulfur-
bond-targeted biodesulfurization. Appl. Environ. Microbiol. 63,
3164–3169.
Li, F.L., Xu, P., Ma, C.Q., Luo, L.L., Wang, X.S., 2003. Deep desulfu-
rization of hydrodesulfurization-treated diesel oil by a facultative
thermophilic bacterium Mycobacterium sp. X7B. FEMS. Micro-
biol. Lett. 223, 301–307.
Li, F.L., Xu, P., Feng, J.H., Meng, L., Zheng, Y., Luo, L.L., Ma, C.Q.,
2005. Microbial desulfurization of gasoline in a Mycobacterium
goodii X7B immobilized-cell system. Appl. Environ. Microbiol.
71, 276–281.
Li, M.Z., Squires, C.H., Monticello, D.J., 1996. Genetic anal-
ysis of the dsz promoter and associated regulatory regions
of Rhodococcus erythropolis IGTS8. J. Bacteriol. 178, 6409–
6418.
Monticello, D.J., 2000. Biodesulfurization and the upgrading of
petroleum distillates. Curr. Opin. Biotechnol. 11, 540–546.
Ohshiro, T., Suzuki, K., Izumi, Y., 1996. Regulation of dibenzoth-
iophene degrading enzyme of Rhodococcus erythropolis D-1. J.
Ferment. Bioeng. 81, 121–124.
Oldfield, C., Pogrebinsky, O., Simmonds, J., Olson, E.S., Kulpa,
C.F., 1997. Elucidation of the metabolic pathway for dibenzothio-
phene desulphurization by Rhodococcus sp. strain IGTS8 (ATCC
53968). Microbiology 143, 2961–2973.
Rhee, S.K., Chang, J.H., Chang, Y.K., Chang, H.N., 1998. Desul-
furization of dibenzothiophene and diesel oils by a newly iso-
lated Gordona strain, CYKS1. Appl. Environ. Microbiol. 64,
2327–2331.
Setti, L., Rossi, M., Lanzarini, G., Pifferi, P.G., 1992. The effect of
n-alkanes in the degradation of dibenzothiophene and of organic
sulfur compounds in heavy oil by a Pseudomonas sp. Biotechnol.
Lett. 14, 515–520.
Setti, L., Lanzarini, G., Pifferi, P.G., 1995. Dibenzothiophene
biodegradation by a Pseudomonas sp. in model solutions. Pro-
cess. Biochem. 30, 721–728.
Tan, T.D., 1995. The logging technique for residual oil in water drive
field. China Offshore Oil Gas (Geol.) 9, 421–427.
Zhou, Z.Y., Zhang, K., 2004. Development situation and prospect of
oil fields in China. Petrol. Explor. Dev. 31, 84–87.