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.