See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/26780195
Characterization and phylogenetic analysis of
biodemulsifier-producing bacteria
Article in Bioresource Technology · September 2009
DOI: 10.1016/j.biortech.2009.07.086 · Source: PubMed
CITATIONS
READS
25
185
6 authors, including:
Xiang-Feng Huang
Jia Liu
54 PUBLICATIONS 349 CITATIONS
43 PUBLICATIONS 293 CITATIONS
Tongji University
SEE PROFILE
Tongji University
SEE PROFILE
Li-Jun Lu
Jing-Cheng Xu
22 PUBLICATIONS 220 CITATIONS
11 PUBLICATIONS 425 CITATIONS
Tongji University
SEE PROFILE
Tongji University
SEE PROFILE
All content following this page was uploaded by Xiang-Feng Huang on 15 January 2017.
The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
Bioresource Technology 101 (2010) 317–323
Contents lists available at ScienceDirect
Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech
Characterization and phylogenetic analysis of biodemulsifier-producing bacteria
Xiang-Feng Huang *, Wei Guan, Jia Liu, Li-Jun Lu, Jing-Cheng Xu *, Qi Zhou
College of Environmental Science and Engineering, State Key Laboratory of Pollution Control and Resource Reuse, Tongji University, Shanghai 20092, China
a r t i c l e
i n f o
Article history:
Received 5 May 2009
Received in revised form 28 July 2009
Accepted 31 July 2009
Available online 31 August 2009
Keywords:
Biodemulsifier-producing bacteria
Demulsification
Phylogenetic analysis
Biosurfactant
a b s t r a c t
Based on demulsification performance, twenty biodemulsifier-producing strains were isolated from
various environmental sources. Five of them achieved nearly or over 90% of emulsion breaking ratio
within 24 h. With the aid of biochemical and physiological tests and 16S rDNA analysis, these isolates
were classified into eleven genera, in which six genera (Brevibacillus sp., Dietzia sp., Ochrobactrum sp.,
Pusillimonas sp., Sphingopyxis sp. and Achromobacter sp.) were firstly reported as demulsifying strains.
Moreover, with data in this study and other literatures, a phylogenic tree was constructed, showing a
rich diversity of demulsifying bacteria. Half of these bacteria belong to Actinobacterale order, which is
famous for hydrocarbon degradation and biosurfactant biosynthesis. However, some strains in the same
genera differed remarkably in demulsifying capability, surface properties and biochemical and physiological characteristics. This implied the biosynthesis and composition of biodemulsifier were more
complicated than expected.
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
Demulsifier is largely consumed in the treatment of emulsions
generated in the recovery and processing of crude petroleum
(Das, 2001). As water content in these emulsions increases, demulsification becomes more problematic. Most emulsions produced in
oilfields are W/O type and almost can be destabilized by chemically-synthesized demulsifier, which is composed of refractory organic polymers. Due to this feature, the separated water from
demulsification should be treated properly before discharge to
avoid hazardous risk on ambient ecosystem. This thereby greatly
limits the application of chemical demulsifiers.
Recently, biodemulsifier has attracted more and more attention
in research due to their low toxicity and easy biodegradability
(Amezcua-Vega et al., 2007). With the introduction of some specific functional groups, biodemulsifiers are able to perform well
under extreme conditions and surpass chemical demulsifiers in a
wide range of applications. (Desai and Banat, 1997).
Up to now several biodemulsifier-producing strains have been
successfully isolated. Extensive studies were conducted particularly on Nocardia amarae (Cairns et al., 1982), Torulopsis bombicola
(Duvnjak and Kosaric, 1987), Bacillus sp. (Janiyani et al., 1994) and
Micrococcus sp. (Das, 2001). However, these studies mainly concentrated on the isolation of demulsifying strains, the optimization of
cultivation conditions and evaluation of demulsifying capability.
Little light has been shed on the sourcing, distribution and
* Corresponding authors. Tel./fax: +86 021 65982592.
E-mail addresses: hxf@tongji.edu.cn, ly008150@online.sh.cn (X.-F. Huang),
xujick@tongji.edu.cn (J.-C. Xu).
0960-8524/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2009.07.086
characterization of demulsifying strains. On the contrary, biosurfactant-producing bacteria have already been studied regarding
their distribution in nature and microbial diversity. They were
found widely present in petroleum-polluted or undisturbed environment (Bodour et al., 2003). With the aid of molecular biological
analysis means, their phylogenic evolution was revealed. It clearly
indicated that the distribution of biosurfactant-producing bacteria
covered three divisions of Eubacteria phylum and two divisions of
Archaea phylum (Bodour et al., 2003). The results provided a
roadmap to the future research, especially in isolation and identification of new biosurfactant-producing strains as well as the
produced biosurfactants. Although a couple of biosurfactant-producing strains isolated from diesel-contaminated soils exhibited
promising potential in environment remediation (Bento et al.,
2005), it has not been clearly identified which species of biosurfactant-producing bacteria are able to break emulsions. As a special
kind of biosurfactant-producing bacteria, demulsifying strains
were not screened solely by surface activity as biosurfactant-producing bacteria, but according to their demulsification ability. This
definitely results in difference of phylogenetic scope between
these two groups of bacteria. For further exploration of biodemulsifier-producing bacteria, it is necessary and urgent to study their
distribution in environment and microbial diversity so as to
improve the screening efficiency of biodemulsifier-producing
bacteria, to direct genetic modification and to optimize the biosynthesis of demulsifier.
The aim of this paper is to study the microbial diversity and
phylogenic relationship of biodemulsifier-producing bacteria.
Based on previous work (Huang et al., 2009), twenty biodemulsifier-producing strains were identified by 16S rDNA sequencing
318
X.-F. Huang et al. / Bioresource Technology 101 (2010) 317–323
Table 1
Code of tested strains and their sources.
Sources
Strain codea
a
Excess activated sludge
Petroleum-contaminated
soil in oilfields
Oil-bearing produced
water from oilfield
Domestic
wastewater
treatment plant
Produced water
treatment station
in Shengli oilfield
Jiangqiao landfill
leachate treatment
plant
Wastewater treatment
station of Gaoqiao
refinery plant
Jiangsu
oilfield
Shengli
oilfield
Xinjiang
oilfield
Jiangsu
oilfield
Shengli
oilfield
ES-SDK-1
ES-SDK-3
ES-QY-1
ES-QY-2
ES-QY-3
ES-SL-1
ES-SL-2
ES-SL-3
ES-JQ-1
ES-JQ-2
ES-LYC-1
S-JS-1
S-JS-8
S-SL-2
S-SL-3
S-XJ-1
S-XJ-12
W-JS-1
W-SL-1
W-SL-2
The code is composed of source type, location and number. Regarding source type, ES stands for excess activated sludge, S for soil and W for wastewater.
analysis with the aid of biochemical and physiological tests. The results were used to construct a phylogenic tree, together with some
published strains in references.
2. Methods
2.1. Tested strains
From nine environmental samples, twenty demulsifying strains
were isolated in previous work (Huang et al., 2009) as shown in
Table 1 and stored in agar slant at 4 °C.
2.2. Culture medium and cultivation conditions
The stored isolates were revived and then inoculated in 100 mL
broth culture for 72-h cultivation in a rotary shaker (at 150 rpm) at
35 °C. The broth culture contained (L1): beef extract 3.0 g, peptone
10.0 g and NaCl 5.0 g at pH 7.0–7.2. After enrichment, 10 mL of aliquot was transferred into 100 mL of MMSM (modified mineral
salts medium) for another 7-day cultivation, and then the culture
was tested for emulsion breaking ratio, surface tension and cell
hydrophobicity. The composition of MMSM was (L1): NH4NO3
4.0 g, K2HPO4 4.0 g, KH2PO4 6.0 g, MgSO47H2O0.2 g, trace mineral
solution 1 mL at pH 7.0–7.2 with 4% (v/v) liquid paraffin as the sole
carbon source. Trace mineral contains (L1) CaCl22H2O1.0 g, FeSO47H2O 1.0 g and EDTA 1.4 g. In addition, agar broth media is used
in slant store of all isolates. It is prepared by adding 20 g/L of agar
powder in the above-mentioned broth culture.
2.3. Demulsifying capability evaluation
The preparation protocols for paraffin-free whole culture, cellfree culture and cell suspension followed what was described by
Park et al. (2000). After 7-day cultivation, the whole culture was
transferred into a separatory funnel and stayed in stagnant condition for 2 h to allow separation of residual paraffin from water
phase. Paraffin-free whole culture was then obtained. In a
100 mL centrifugal tube, 60 mL of paraffin-free whole culture
was added and then centrifuged at 10,000 rpm for 20 min. The liquid obtained passed through the 0.45 lm membrane filter to
yield cell-free culture. Cell pellets which settled in the bottom of
the centrifugal tube, was rinsed and re-suspended in 60 mL of distilled water as cell suspension.
W/O kerosene emulsion was prepared according to the protocol
in reference (Nadarajah et al., 2002). The emulsion was obtained by
mixing distilled water and kerosene (1:1 by v/v) with the aid of
1.67% (w/v) Span80 at 10,000 rpm for 3 min. The emulsion type
was identified by Oil Red O-test as described by Lee and Lee
(2000). The emulsion showed less than 10% of emulsion breaking
ratio at 35 °C within 24 h.
In a demulsification test, 2 mL of paraffin-free whole culture or
cell-free supernatant or cell suspension was added into a 20 mL
graduated test tube containing 18 mL of model emulsion. The test
tube was vigorously inverted for 200 times to achieve complete
mixing and then left undisturbed in water baths at 35 °C. The
volume of residual emulsion was recorded at certain intervals.
Same amount of sterile culture media and chemically-synthesized
polyether demulsifier were used to substitute biodemulsifier in a
blank test and a comparison test, respectively. The demulsification performance was evaluated by emulsion breaking ratio as
follows:
Emulsion breaking ratio
remaining emulsion volume
¼ 1
original emulsion volume þ added culture volume
100%
Cell-free supernatant and cell suspension of the isolates were
tested separately in demulsification tests to identify the location
of its biodemulsifier. In this study, a strain is regarded as having
demulsifying capability when it achieved over 50% of emulsion
breaking ratio in demulsification test. Similarly, the produced biodemulsifier is regarded as extracellular if the emulsion breaking ratio of cell-free supernatant was more than 50% (within 24 h). On
the contrary, it is reckoned as cell-bound. If both exhibited more
than 50% of demulsification ratio, the biodemulsifier is present in
both cell-wall and supernatant.
2.4. Measurement of surface tension and microbial adhesion to the
hydrocarbon (MATH)
The surface tension of culture was measured by Du Nouy ring
tensiometer (DT-102, Zibo Huakun Electrical Equipment Limited
Company, China). Each observation was reported as the average
of triplicate measurements.
The microbial adhesion to the hydrocarbon (MATH) was used to
denote the surface cell hydrophobicity of the microbial species. It
was measured according to the protocol suggested by Thavasi
et al. (2008). The cell pellets obtained after centrifuge was rinsed
with phosphate buffer pH 7.0 twice and then diluted to obtain initial OD580 of around 0.8–1.0. Then 5 ml of this cell suspension was
mixed with same volume of kerosene in a tube on a vortexer at
2500 rpm for 5 min. The mixture was left undisturbed for 5 min,
and then final OD580 of the aqueous phase was measured. MATH
was calculated as follows:
MATH ¼ 1 OD580ðfinalÞ =OD580ðinitialÞ 100%
This value varies from 0% to 100%. A higher value indicates
higher cell hydrophobicity.
X.-F. Huang et al. / Bioresource Technology 101 (2010) 317–323
2.5. Biochemical and physiological characterization
The biochemical and physiological characterization of each isolate followed Bergey’s Manual (the 9th edition) (Holt et al., 1994).
For various species, different tests were conducted, including Gram
staining, acid-fast staining, spore staining, oxidase reaction, catalase reaction, glucose fermentation, acid production (from glucose,
xylose, maltose and mannitol), nitrate reduction, hydrogen sulfide
production, indole production, enzyme production (such as
b-galactosidase, lysine-decarboxylase, ornithine decarboxylase,
acetamidase, urease, arginine dihydrolase), MacConkey cultivation,
citrate utilization and Voges Proskauer (VP) reaction.
2.6. PCR and sequencing identification of 16S rDNA and construction of
phylogenic tree
In the identification of 16S rDNA, DNA was extracted with DNA
isolation kit (Omega). 16S rDNA encoding genes were amplified by
PCR using universal primer 8F (50 -AGAGTTTGATCCTGGCTCAG-3’)
and 1492R (5’-GGTTACCTTGTTACGACTT-30 ). Each 50 lL of PCR
mixture contained 1.5 lL template DNA, 0.25 lL Taq DNA polymerase, 5 lL PCR buffer, 5 lL 2.5 mM dNTPs, 2 lL of each primer
(6.5 lmol/L) and double distilled water. PCR program for amplification was initial denaturation at 94 °C for 3 min followed by 35
cycles consisting of 94 °C for 30 s, 56 °C for 30 s and 72 °C for 80
s, and a final extension at 72 °C for 10 min. The PCR product was
visualized on agarose gel by electrophoresis. The purified PCR
product was ligated to pMD18-T vector and then transferred into
Escherichia coli Top10 for cultivation. When colonies developed in
agar plate, PCR identification was carried out with universal primer
M13R(50 -AGCGGATAACAATTTCACACAGGA-30 ) and M13F(50 -CGCCAGGGTTTTCCCAGTCACGAC-30 ) for T vectors in the same procedure
as described above. The sequence of cloned product was analyzed
by Invitrogen.
The 16S rDNA gene nucleotide sequence of each isolate was
compared with known sequences in GenBank using BLAST program. The sequence alignment was carried out with ClusterX
(version 2.0). Sequences of isolates from this studies and other
references were used to construct a phylogenic trees with PAUP*
(version 4.0) by using neighbor-jointing method with HasegawaKishino-Yano (HKY85) model (Bodour et al., 2003).
3. Results and discussion
3.1. Demulsifying capability and surface properties of tested strains
The paraffin-free culture of each isolate was examined for surface tension, cell hydrophobicity, location of produced biodemulsifier and emulsion breaking. The results were shown in Table 2.
All tested strains showed more than 50% of emulsion breaking
ratio, in which five strains (ES-QY-3, ES-SL-1, S-SL-2, S-XJ-1, S-XJ12) exhibited nearly or over 90% of emulsion breaking ratio, much
higher than the ratio achieved by the chemical demulsifier
(73 ± 2%). The performance of the rest was either similar or inferior
to the chemical demulsifier. In previous studies, the production
and type of the produced biosurfactant was found closely related
to carbon source and other cultivation conditions (Gogotov and
Khodakov, 2008). Paraffin is the sole carbon source in this study.
Various strains may have different preference in carbon sources
and different utilization rate of paraffin. Thereby, the demulsifying
capability of those inferior strains would be improved by changing
the carbon source or optimizing cultivation conditions.
Biodemulsifiers are a group of surface-active bio-products with
both hydrophilic and hydrophobic ends. For example, with aid of
thin layer chromatography and Fourier transform infrared
319
spectrometer, the biodemulsifier produced by S-XJ-1 was identified as lipopeptide, which had amino-acid chain as hydrophilic
end and hydrocarbon chain as hydrophobic end (Huang et al.,
2009). A good demulsifier possesses three traits: good miscibility
in dispersed phase, good diffusivity to ensure a high enough diffusion flux to the interface and high surface activity to reduce the
gradients of interfacial tension (Ivanov and Kralchevsky, 1997).
The amphipathic molecule of biodemulsifier facilitates its adsorption on oil/water interface and helps to alter the properties of the
interface. It was found in demulsification tests that the cultures
of five highly-efficient strains quickly disperse into the emulsion,
showing good miscibility. Moreover, their culture all displayed
high surface activity with surface tension ranging from 30 to
50 mN/m. In an emulsion, this feature enables the biodemulsifier
to substitute emulsifier at the interface, which further leads to flocculation and coalescence of dispersed globules and eventually
complete separation of two phases. Besides the amphipathic molecule of biodemulsifer, high cell hydrophobicity can also contribute to demulsification to some extent. For example, four of the
five strains exhibited a high MATH value, which indicated a high
affinity of the cells to kerosene.
The location of produced biodemulsifer determined the degree
of difficulty in its extraction and purification, which not only affected the in-depth investigation of biodemulsifier, but also contributed significantly to the production cost. As a result, research
advances in a quicker pace for extracellular biodemulsifier, which
are readily-extractable. As shown in Table 2, nine strains produced
extracellular biodemulsifier, eight produced cell-bound and three
produced both. Moreover, highly-efficient biodemulsifier (with
over 90% of emulsion breaking ratio) can be found in both extracellular and cell-bound form. Simultaneous production of extracellular and cell-bound biosurfactants were also reported in other
research with hydrocarbon as substrate (Franzetti et al., 2008).
By comparing the sources of these strains, it was found that most
of the strains isolated from activated sludge produced extracellular
biodemulsifiers, while the strains from petroleum-contaminated
soils and waters mainly synthesized cell-bound biodemulsifiers.
In this study, the strains were isolated from samples taken from
several oilfields, which scattered over China, and wastewater treatment plants treating various wastewater. They represented a variety of sources (as shown in Table 1). It is shown in Table 2 that
surface properties of the tested isolates showed some correlation
with their sources. Seven strains with low surface tension and nine
strains with high MATH were isolated from petroleum-related
sources, such as produced water of oilfield, activated sludge treating produced water or oilfield soils. Moreover, four of the five
highly-efficient demulsifying strains (except ES-QY-3) were also
isolated from petroleum-related sources. This may be explained
as follows. In petroleum-contaminated environment, microbial
diversity decreases due to the natural selection and only strains
capable of utilizing bio-toxic hydrocarbons will survive and predominate (Li et al., 2000). To facilitate the utilization of hydrocarbons, these strains would either produce surface-active substances
or increase their affinity to oil by altering cell hydrophobicity.
3.2. Identification of tested strains and phylogenic study
All these isolates were chemoheterotrophs with four Gram-positive and sixteen Gram-negative. The morphology of cells and colonies as well as their biochemical and physiological characteristics
were listed in Electronic Annex 1 in the online version of this article.
Due to intrinsic limitations, the biochemical and physiological
features can only provide preliminarily identification (Bizet et al.,
1997). The final identification of strains was accomplished by
combing the alignment results of 16S rDNA sequence analysis with
biochemical and physiological characteristics.
320
X.-F. Huang et al. / Bioresource Technology 101 (2010) 317–323
Table 2
Surface properties and demulsifying capability of tested isolatesa.
Source
No.
Strain code
Paraffin-free culture
Emulsion breaking ratio
in 24 h(%)b
Excess sludge
1
2
3
4
5
6
7
8
9
10
11
ES-JQ-1
ES-JQ-2
ES-LYC-1
ES-QY-1
ES-QY-2
ES-QY-3
ES-SDK-1
ES-SDK-3
ES-SL-1
ES-SL-2
ES-SL-3
Petroleum-contaminated
soil
12
13
14
15
16
17
Oil-bearing produced water
from oilfield
18
19
20
MATH (%)
Location of produced
biodemulsifierc
Surface tension
(mN/m)
65 ± 5
68 ± 1
70 ± 2
69 ± 2
50 ± 4
98 ± 2
59 ± 4
62 ± 3
99 ± 1
58 ± 5
67 ± 4
42.6 ± 0.4
45.8 ± 0.3
38.8 ± 0.5
41.1 ± 0.2
49.1 ± 0.9
46.3 ± 0.3
49.3 ± 0.5
48.1 ± 0.4
49.3 ± 0.3
54.2 ± 1.7
54.8 ± 2.2
44 ± 6
8±6
54 ± 5
68 ± 1
28 ± 4
17 ± 0
20 ± 1
0.3 ± 0
57 ± 2
9±1
65 ± 4
E
E
E&C
E
E
E
E
E
E
C
E
S-JS-1
S-JS-8
S-SL-2
S-SL-3
S-XJ-1
S-XJ-12
72 ± 2
65 ± 3
88 ± 3
50 ± 2
91 ± 4
100 ± 0
30.3 ± 0.0
36.0 ± 1.9
34.6 ± 1.3
32.8 ± 0.2
33.9 ± 1.1
30.4 ± 0.2
70 ± 6
70 ± 1
81 ± 2
50 ± 2
62 ± 1
46 ± 0
C
C
E&C
E&C
C
C
W-JS-1
W-SL-1
W-SL-2
50 ± 2
56 ± 4
68 ± 3
53.7 ± 0.2
54.4 ± 0.5
50.5 ± 1.4
54 ± 4
26 ± 0
14 ± 4
C
C
C
a
Data were reported as average of triplicate observations ± standard deviation.
The blank (with 2 ml of virgin culture instead of biodemulsifier) showed 3 ± 1% of emulsion braking ratio within 24 h.
c
As to the location of biodemulsifier, E stands for extracellular and C for cell-bound. It was determined according to what is described in Section 2.3. Emulsion breaking
ratios of cell-free supernatant and cell suspension are not shown here.
b
Strains ES-SDK-3 exhibited high homology (99%) with Castellaniella sp. and Alcaligenes sp. Castellaniella sp. is capable of degrading
nonylphenol ethoxylates (NPnEO) with low ethoxylation degree,
which is particularly recalcitrant to biodegradation (Di Gioia
et al., 2008). But no reports can be found on the demulsifying capability of this genus. Although it was reported as heterocyclic
amine-degrading denitrifying bacteria, Alcaligenes sp. was also
found showing surface activity (producing bioemulsan) (Toledo
et al., 2008) or demulsifying capability (Nadarajah et al., 2002).
The phylogenetic analysis (the phylogenic tree were not shown
here) indicated that ES-SDK-3 was closer to Alcaligenes sp. and thus
it is probably a species of Alcaligenes sp.
Despite 98% of homology with both Castellaniella sp. and Alcaligenes sp., strain ES-JQ-2 was also classified into the latter genus in a
similar way.
The highest homology obtained by strain S-JS-1 was 93% of similarity to Dietzia sp., which, together with Rhodococcus sp., belongs
to Nocardioform Actinomycetes (Holt et al., 1994). Both showed similar features (Bizet et al., 1997). Previous studies showed that Rhodococcus sp. was able to produce biosurfactant (Gogotov and
Khodakov, 2008) and break emulsions (Das, 2001). Up to now, Dietzia sp. was mainly isolated from petroleum-polluted soils and reported in degradation of oils (von der Weid et al., 2007). Because
the biochemical and physiological features of strain S-JS-1 complied with those reported for Dietzia sp., it is probably a new species of Dietzia sp.
The rest seventeen strains all obtained single alignment result
with more than 97% of similarity to the assigned genus. Their morphological, biochemical and physiological features were also consistent with the description in references. If 97% similarity of 16S
rDNA gene sequence is defined as a good match standard, nineteen
of the twenty tested strains can be identified into eleven genera as
show in Table 3.
3.3. Phylogenic analysis of biodemulsifier-producing bacteria
With the addition of the eleven genera found in this study, all
the known demulsifying bacteria can be classified into the follow-
ing genera, Nocardia sp.(Stewart et al., 1983), Corynebacterium
sp.(Stewart et al., 1983), Bacillus sp.(Janiyani et al., 1994), Rhodococcus sp.(Ma et al., 2006; Wilkinson and Cooper, 1985), Alcaligenes
sp.(Nadarajah et al., 2002), Pseudomonas sp.(Nadarajah et al.,
2002), Streptomyces sp.(Park et al., 2000), Micrococcus sp.(Das,
2001), Torulopsis bombicola (Duvnjak and Kosaric, 1987), Mycobacterium sp.(Nadarajah et al., 2002), Acinetobacer sp.(Nadarajah et al.,
2002), Alteromonas sp.(Nishimaki et al., 1999), Aeromonas
sp.(Nishimaki et al., 1999) and Arthrobacter sp.(Das, 2001). In order
to reveal the evolution of demulsifying bacteria, the 16S rDNA sequence of all found strains were used to construct a phylogenic
tree with PAUP by using the neighbor-joining method.
The phylogenic tree (as shown in Fig. 1) indicated a rich diversity of demulsifying bacteria, which covered three divisions (Proteobacteria, Firmicutes and Actinobacteria) of Eubacteria and
Ascomycota division of Eukarya. It is worth noting that the classification of some strains published long time ago has been updated
due to recent advances in bacterial taxonomy. For example, N.
amarae ATCC72808 (Stewart et al., 1983) has been re-classified
into Gordonia amarae according to the analysis result of small-subunit ribosomal DNA sequence (Ruimy et al., 1994). Another species,
Rhodococcus aurantiacus, which was re-classified into Gordona aurantiaca, was then grouped into Tsukamurella inchonensis and has
recently been changed to Tsukamurella paurometabola by Kattar
et al. (2001). Other examples are Rhodococcus rubropertinctus reclassified into Gordonia rubripertinctu (Rainey et al., 1995); Corynebacterium fascians re-grouped into Rhodococcus fascians in 1984
(Rainey et al., 1995); a demulsifying yeast, Torulopsis bombicola
ATCC22214 (Duvnjak and Kosaric, 1987) re-classified into Candida
bombicola (Odds et al., 1997). As a result, demulsifying strains are
expanded to three genera (Gordonia sp., Tsukamurella sp. and Candida sp.), which still belong to the four divisions mentioned above.
The isolates in this study were evenly distributed among Proteobacteria, Firmicutes and Actinobacteria division. So far no demulsifying strain was found in Fungus. Except five genera (including two
of Rhodococcus sp., two of Bacillus sp., three of Alcaligenes sp., two of
Gordonia sp. and one of Pseudomonas sp.) were already reported in
previous biodemulsifier studies, the remaining six genera of this
321
X.-F. Huang et al. / Bioresource Technology 101 (2010) 317–323
Table 3
Identification of the twenty tested strains with 16S rDNA gene analysis.
No.
Strain code
GenBank
Register No.
16S rDNA gene sequence
Genera
Nearest relative in GenBank
Similarity (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
ES-JQ-1
ES-JQ-2
ES-LYC-1
ES-QY-1
ES-QY-2
ES-QY-3
ES-SDK-1
ES-SDK-3
ES-SL-1
ES-SL-2
ES-SL-3
S-JS-1
S-JS-8
S-SL-2
S-SL-3
S-XJ-1
S-XJ-12
W-JS-1
W-SL-1
W-SL-2
FJ529026
FJ529027
FJ529028
FJ529029
FJ529030
FJ529031
FJ529037
FJ531636
FJ529038
FJ529039
FJ529040
FJ529032
FJ529034
FJ529022
FJ529024
EF117974
FJ529035
FJ529036
FJ529041
FJ529042
Brevibacillus agri strain NCHU1002(AY319301)
Alcaligenes sp. strain Ic4(DQ421393)
Gordonia sp. strain D2(DQ787430)
Dietzia sp. strain ES18(AY028326)
Ochrobactrum intermedium clone kl-2(EU816698)
Pusillimonas terrae strain BN9T(DQ466075)
Sphingopyxis granuli strain Kw07(AY563034)
Alcaligenes sp. strain ESPY2(A-III)(EF205261)
Brevibacillus borstelensis strain MH301(DQ350830)
Bacillus badius strain NBRC 15713(AB271748)
Brevibacillus borstelensis strain T2–1(AB215102)
Dietzia natronolimnaea strain LL 51(DQ821754)
Pseudomonas sp. strain BFXJ-8(EU013945)
Rhodococcus sp. strain F12(EU697084)
Rhodococcus sp. strain E33(AY114109)
Alcaligenes sp. strain mp-2(AY331576)
Dietzia sp. strain ES18(AY028326)
Gordonia sp. strain D2(DQ787430)
Achromobacter sp. strain EP17(AM398226)
Bacillus sp. strain BSi20511(EF673289)
99
98
99
100
99
98
99
99
99
99
99
93
99
99
100
99
100
99
99
99
Brevibacillus sp.
Alcaligenes sp.
Gordonia sp.
Dietzia sp.
Ochrobactrum sp.
Pusillimonas sp.
Sphingopyxis sp.
Alcaligenes sp.
Brevibacillus sp.
Bacillus sp.
Brevibacillus sp.
Dietzia sp.
Pseudomonas sp.
Rhodococcus sp.
Rhodococcus sp.
Alcaligenes sp.
Dietzia sp.
Gordonia sp.
Achromobacter sp.
Bacillus sp.
Fig. 1. Phylogenic tree was based on gene sequences, showing the position of 20 demulsifying strains found in this study and published demulsifying strains from literatures.
The unrooted tree was created by using the neighbor-joining method. The bar indicates 0.05 substitutions per nucleotide position.
study were firstly reported as demulsifying strains. They were
Ochrobactrum sp., Pusillimonas sp., Sphingopyxis sp. and Achromobacter sp. of Proteobacteria division; Brevibacillus sp. of Firmicutes
division and Dietzia sp. of Actinobacteria division. In addition, this
study has contributed five demulsifying strains (ES-JQ-2, ES-QY3, ES-SDK-3, S-XJ-1, W-SL-1) to Betaproteobacteria class. This
greatly increased the quantity of demulsifying strains in this class.
As shown by the phylogenic tree, more than ten demulsifying
genera were found in Actinobacteria class of Actinobacteria division
(according to LPSN and NCBI taxonomy system). In this study, five
of seven isolates in this division (ES-LYC-1, S-JS-1, S-SL-2, S-SL-3,
S-XJ-12) were able to reduce surface tension below 40 mN/m; six
(ES-LYC-1, ES-QY-1, S-JS-1, S-SL-2, S-SL-3, W-JS-1) showed more
than 50% of MATH and two achieved approximately 90% of
emulsion breaking ratio. Compared with strains in the other two
divisions, most of the strains in Actinobacteria division showed
higher surface activity and cell hydrophobicity, indicating promising potential in biodegradation of hydrocarbon and production of
biosurfactant. As a result, strains in Actinobacteridae – Actinobacterale order also exhibited other valuable capabilities such as
biosynthesis of antibiotic and useful metabolism byproducts,
which can be applied in biological treatment of wastewater or solid
wastes (Goodfellow et al., 1996). For instance, Mycobacterium sp.,
Corynebacterium sp. and Brevibacterium sp. were able to produce
extracellular lipopeptide biosurfactants (Haferburg et al., 1986)
and a Rhodococcus wratislaviensis strain BN38 was capable of
322
X.-F. Huang et al. / Bioresource Technology 101 (2010) 317–323
producing glycolipid biosurfactants (Tuleva et al., 2008). Some
strains, such as Mycobacterium, Rhodococcus, Dietzia and Gordonia
alkanivorans, exhibited ability of degrading hydrocarbons (Brito
et al., 2006; Ta-Chen et al., 2008).
In this study, six of eleven genera contained more than two
strains. Surprisingly, the strains in the same genera were distinct
in many aspects. For example, both ES-LYC-1 and W-JS-1 were
identified as Gordonia sp., and both showed 99% of homology in
16S rDNA analysis. However, they gave opposite response in gram
staining, oxidase reaction and hydrogen sulfide production.
Regarding demulsification performance, W-JS-1 performed far
poorer than ES-LYC-1 (in Table 2). The surface tension of their cultures also differed remarkably. Strain ES-LYC-1 produced both
extracellular and cell-bound biodemulsifier, while W-JS-1 only
produced cell-bound biodemulsifier. Distinctions were also observed in other genera, such as ES-QY-1 and S-XJ-12 of Dietzia
sp., even though these two strains showed 100% of homology in
their 16S rDNA sequence. As reported in previous study, the
homology mixture of biosurfactant produced by strains in the
same genera may differed slightly in the composition or chemical
structures, which would have great consequence on its surface
activity (Franzetti et al., 2008). Similar to biosurfactant, our previous work also showed that produced biodemulsifer was a mixture
of homologue (data not shown). Moreover, it was found that two
strains displayed different surface activity, even though there
was a difference of only 1 bp in the 1466 bp of their 16S rRNA
sequence (Bodour et al., 2003). This suggested that more sophisticated molecular approaches, such as repetitive extragenic
palindromic (REP)-PCR fingerprint should be used to further differentiate these strains.
Comparing the sources of each isolate, it can be seen that
eleven strains belonging to nine genera were isolated from excess
activated sludge; three isolates classified into three genera were
from oil-bearing wastewater and six isolates grouped into four
genera were from petroleum-contaminated soils. Moreover,
strains isolated from activated sludge and oil-bearing wastewater
covered Proteobacteria, Actinobacteria and Firmicutes division
while strains from petroleum-polluted soils only distributed in
the former two divisions. Compared with the other two sources,
strain from activated sludge showed a richer microbial diversity.
This may be explained by rich composition of treated wastewaters, especially diverse substrates and abundant nutrients, which
provides a favorable environment for the survival of diverse
microbes.
4. Conclusions
In this study, twenty demulsifying strains were isolated from
environmental sources. Five of them were highly-efficient in
demulsification, showing nearly or over 90% of emulsion breaking
ratio. By combining the results of this study with previous findings,
a phylogenic tree of demulsifying bacteria was created. It showed
that all the known demulsifying strains were classified into eleven
genera and distributed among Proteobacteria, Firmicutes and
Actinobacteria division in Eubacteria and Ascomycota division in
Eucaryote. The findings of this study added six new genera to biodemulsifier-producing bacteria. Strains isolated from activated
sludge showed a wider diversity, including nine genera of three
divisions.
Acknowledgements
The authors would like to express their gratitude to Science and
Technology Commission of Shanghai Municipality (Grant No.
06DZ22007), and Science and Technology Agency of Xinjiang Uy-
gur Autonomous Region (Grant No. 200891115) for the financial
support.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.biortech.2009.07.086.
References
Amezcua-Vega, C., Poggi-Varaldo, H.M., Esparza-Garcia, F., Rios-Leal, E., RodriguezVazquez, R., 2007. Effect of culture conditions on fatty acids composition of a
biosurfactant produced by Candida ingens and changes of surface tension of
culture media. Bioresource Technology 98, 237–240.
Bento, F.M., Camargo, F.A.D., Okeke, B.C., Frankenberger, W.T., 2005. Diversity of
biosurfactant producing microorganisms isolated from soils contaminated with
diesel oil. Microbiological Research 160, 249–255.
Bizet, C., Barreau, C., Harmant, C., Nowakowski, M., Pietfroid, A., 1997. Identification
of Rhodococcus, Gordona and Dietzia species using carbon source utilization tests
(‘‘Biotype-100” strips). Research in Microbiology 148, 799–809.
Bodour, A.A., Drees, K.P., Maier, R.M., 2003. Distribution of biosurfactant-producing
bacteria in undisturbed and contaminated arid southwestern soils. Applied and
Environmental Microbiology 69, 3280–3287.
Brito, E.M., Guyoneaud, R., Goni-Urriza, M., Ranchou-Peyruse, A., Verbaere, A.,
Crapez, M.A.C., Wasserman, J.C.A., Duran, R., 2006. Characterization of
hydrocarbonoclastic bacterial communities from mangrove sediments in
Guanabara Bay, Brazil. Research in Microbiology 157, 752–762.
Cairns, W.L., Cooper, D.G., Zajic, J.E., Wood, J.M., Kosaric, N., 1982. Characterization
of Nocardia amarae as a potent biological coalescing agent of water–oil
emulsions. Applied Environmental Microbiology 43, 362–366.
Das, M., 2001. Characterization of de-emulsification capabilities of a Micrococcus
species. Bioresource Technology 79, 15–22.
Desai, J.D., Banat, I.M., 1997. Microbial production of surfactants and their
commercial potential. Microbiology and Molecular Biology Reviews 61, 47–64.
Di Gioia, D., Michelles, A., Pierini, M., Bogialli, S., Fava, F., Barberio, C., 2008. Selection
and characterization of aerobic bacteria capable of degrading commercial
mixtures of low-ethoxylated nonylphenols. Journal of Applied Microbiology
104, 231–242.
Duvnjak, Z., Kosaric, N., 1987. De-emulsification of petroleum water-in-oil
emulsions by selected bacterial and yeast cells. Biotechnology Letters 9, 39–42.
Franzetti, A., Bestetti, G., Caredda, P., La Colla, P., Tamburini, E., 2008. Surface-active
compounds and their role in the access to hydrocarbons in Gordonia strains.
FEMS Microbiology Ecology 63, 238–248.
Gogotov, I.N., Khodakov, R.S., 2008. Surfactant production by the Rhodococcus
erythropolis sH-5 bacterium grown on various carbon sources. Applied
Biochemistry and Microbiology 44, 186–191.
Goodfellow, M., Davenport, R., Stainsby, F.M., Curtis, T.P., 1996. Actinomycete
diversity associated with foaming in activated sludge plants. Journal of
Industrial Microbiology 17, 268–280.
Haferburg, D., Hommel, R., Claus, R., Kleber, H.-P., 1986. Extracellular microbial
lipids as biosurfactants. In: Fiechter, A. (Ed.), Advances in Biochemical
Engineering/Biotechnology. Springer-Verlag, Berlin, Heidelberg, pp. 53–93.
Holt, J.G., Krieg, N.R., Sneath, P.H.A., Staley, J.T., Williams, S.T., 1994. Bergey’s Manual
of Determinative Bacteriology, ninth ed. Lippincott, Williams & Wilkins,
Philadelphia, PA 19106, USA.
Huang, X.F., Liu, J., Lu, L.J., Wen, Y., Xu, J.C., Yang, D.H., Zhou, Q., 2009. Evaluation of
screening methods for demulsifying bacteria and characterization of
lipopeptide bio-demulsifier produced by Alcaligenes sp. Bioresource
Technology 100, 1358–1365.
Ivanov, I.B., Kralchevsky, P.A., 1997. Stability of emulsions under equilibrium and
dynamic conditions. Colloids and Surfaces A: Physicochemical and Engineering
Aspects 128, 155–175.
Janiyani, K.L., Purohit, H.J., Shanker, R., Khanna, P., 1994. De-emulsification of oil-inwater emulsions by Bacillus-subtilis. World Journal of Microbiology and
Biotechnology 10, 452–456.
Kattar, M.M., Cookson, B.T., Carlson, L.C., Stiglich, S.K., Schwartz, M.A., Nguyen, T.T.,
Daza, R., Wallis, C.K., Yarfitz, S.L., Coyle, M.B., 2001. Tsukamurella strandjordae
sp. nov., a proposed new species causing sepsis. Journal of Clinical Microbiology
39, 1467–1476.
Lee, J.C., Lee, K.Y., 2000. Emulsification using environmental compatible emulsifiers
and de-emulsification using DC field and immobilized Nocardia amarae.
Biotechnology Letters 22, 1157–1163.
Li, G.H., Zhang, X., Huang, W., 2000. Characteristics of degrading microorganism
distribution in polluted soil with petroleum hydrocarbons. Huanjing Kexue 21,
61–64.
Ma, T., Ling, F.-L., Xi, Y.-W., Liu, R.-L., 2006. Performance of demulsification by
Rhodococcus sp.PR-1. Huanjing Kexue 27, 1191–1196.
Nadarajah, N., Singh, A., Ward, O.P., 2002. Evaluation of a mixed bacterial culture for
de-emulsification of water-in-petroleum oil emulsions. World Journal of
Microbiology and Biotechnology 18, 435–440.
Nishimaki, F., Takahashi, N., Tsuchida, T., Watanabe, K., Hino, S., 1999.
Microorganisms, demulsifiers and processes for breaking an emulsion. In:
X.-F. Huang et al. / Bioresource Technology 101 (2010) 317–323
FreePatentsOnline/PatentStorm (Ed.), US Patent Office. Tonen Corporation
(Tokyo, JP), USA.
Odds, F.C., Rinaldi, M.G., Cooper, C.R., Fothergill, A., Pasarell, L., McGinnis, M.R.,
1997. Candida and Torulopsis: a blinded evaluation of use of pseudohypha
formation as basis for identification of medically important yeasts. Journal of
Clinical Microbiology 35, 313–316.
Park, S.H., Lee, J.H., Ko, S.H., Lee, D.S., Lee, H.K., 2000. Demulsification of oil-in-water
emulsions by aerial spores of a Streptomyces sp. Biotechnology Letters 22, 1389–
1395.
Rainey, F.A., Burghardt, J., Kroppenstedt, R.M., Klatte, S., Stackebrandt, E., 1995.
Phylogenetic analysis of the genera Rhodococcus and Nocardia and evidence for
the evolutionary origin of the genus Nocardia from within the radiation of
Rhodococcus Species. Microbiology-UK 141, 523–528.
Ruimy, R., Boiron, P., Boivin, V., Christen, R., 1994. A Phylogeny of the genus
Nocardia deduced from the analysis of small-subunit ribosomal DNAsequences, including transfer of Nocardia-amarae to the genus Gordona as
Gordona-amarae comb-nov. FEMS Microbiology Letters 123, 261–267.
Stewart, A.L., Gray, N.C.C., Cairns, W.L., Kosaric, N., 1983. Bacteria-induced deemulsification of water-in-oil petroleum emulsions. Biotechnology Letters 5,
725–730.
View publication stats
323
Ta-Chen, L., Chang, J.S., Young, C.C., 2008. Exopolysaccharides produced by Gordonia
alkanivorans enhance bacterial degradation activity for diesel. Biotechnology
Letters 30, 1201–1206.
Thavasi, R., Jayalakshmi, S., Balasubramanian, T., Banat, I.M., 2008. Production and
characterization of a glycolipid biosurfactant from Bacillus megaterium using
economically cheaper sources. World Journal of Microbiology and
Biotechnology 24, 917–925.
Toledo, F.L., Gonzalez-Lopez, J., Calvo, C., 2008. Production of bioemulsifier by
Bacillus subtilis, Alcaligenes faecalis and Enterobacter species in liquid culture.
Bioresource Technology 99, 8470–8475.
Tuleva, B., Christova, N., Cohen, R., Stoev, G., Stoineva, I., 2008. Production and
structural elucidation of trehalose tetraesters (biosurfactants) from a novel
alkanothrophic Rhodococcus wratislaviensis strain. Journal of Applied
Microbiology 104, 1703–1710.
von der Weid, I., Marques, J.M., Cunha, C.D., Lippi, R.K., dos Santos, S.C.C., Rosado,
A.S., Lins, U., Seldin, L., 2007. Identification and biodegradation potential of a
novel strain of Dietzia cinnamea isolated from a petroleum-contaminated
tropical soil. Systematic and Applied Microbiology 30, 331–339.
Wilkinson, M.A., Cooper, D.G., 1985. Testing of microbial demulsifiers with heavy
crude emulsions. Biotechnology Letters 7, 406–408.