Geobacillus zalihae sp. nov., a thermophilic lipolytic bacterium isolated from palm oil mill effluent in Malaysia

Background Thermophilic Bacillus strains of phylogenetic Bacillus rRNA group 5 were described as a new genus Geobacillus. Their geographical distribution included oilfields, hay compost, hydrothermal vent or soils. The members from the genus Geobacillus have a growth temperatures ranging from 35 to 78°C and contained iso-branched saturated fatty acids (iso-15:0, iso-16:0 and iso-17:0) as the major fatty acids. The members of Geobacillus have similarity in their 16S rRNA gene sequences (96.5–99.2%). Thermophiles harboring intrinsically stable enzymes are suitable for industrial applications. The quest for intrinsically thermostable lipases from thermophiles is a prominent task due to the laborious processes via genetic modification. Results Twenty-nine putative lipase producers were screened and isolated from palm oil mill effluent in Malaysia. Of these, isolate T1T was chosen for further study as relatively higher lipase activity was detected quantitatively. The crude T1 lipase showed high optimum temperature of 70°C and was also stable up to 60°C without significant loss of crude enzyme activity. Strain T1T was a Gram-positive, rod-shaped, endospore forming bacterium. On the basic of 16S rDNA analysis, strain T1T was shown to belong to the Bacillus rRNA group 5 related to Geobacillus thermoleovorans (DSM 5366T) and Geobacillus kaustophilus (DSM 7263T). Chemotaxonomic data of cellular fatty acids supported the affiliation of strain T1T to the genus Geobacillus. The results of physiological and biochemical tests, DNA/DNA hybridization, RiboPrint analysis, the length of lipase gene and protein pattern allowed genotypic and phenotypic differentiation of strain T1T from its validly published closest phylogenetic neighbors. Strain T1T therefore represents a novel species, for which the name Geobacillus zalihae sp. nov. is proposed, with the type strain T1T (=DSM 18318T; NBRC 101842T). Conclusion Strain T1T was able to secrete extracellular thermostable lipase into culture medium. The strain T1T was identified as Geobacillus zalihae T1T as it differs from its type strains Geobacillus kaustophilus (DSM 7263T) and Geobacillus thermoleovorans (DSM 5366T) on some physiological studies, cellular fatty acids composition, RiboPrint analysis, length of lipase gene and protein profile.

The stability of biocatalysts is an important criterion when dealing with bioprocesses at high temperature for sustainable operation. Enzyme stability is dictated by its threedimensional configuration, which in turn is determined by genetic and environmental factors [32]. Therefore, thermophiles are promising sources of heat-stable enzymes. In addition to higher thermostability, proteins from thermophiles often showed higher stability toward organic solvents and higher activity at elevated temperature [33]. In addition, genetic engineering in altering the stability of enzymes is a difficult task and laborious processes. Therefore, efforts have been focused on the screening of microorganisms harboring intrinsically stable biocatalysts. Putative lipase producers were screened quantitatively across various basal media to check for lipase production and the stability of lipase. The highest lipase producing strain have been identified and characterized intensively. Screening and isolation of heat-stable lipase producers are important to fulfill industrial requirements for the desired characteristics. Identification of industrially important enzyme producer is conducted to determine its phylogenetic position in systematic microbiology.

Screening and isolation of thermophilic lipolytic bacteria
Rubbish dump sites and palm oil mill effluent are potential sites containing thermophilic lipolytic bacteria because these sites served as sewage discharge areas for household waste and palm oil factory. The oily environment may provide a good environment for lipolytic microorganisms to flourish. Samples were collected from the rubbish dump site and palm oil mill effluent at different treatment areas. Samples were enriched with enrichment medium (EM-1) containing olive oil as the sole carbon source at 60°C for 2 days under shaking condition to promote the growth of thermophilic lipolytic bacteria. Twenty nine putative lipase producers gave positive results on triolein agar plate by forming an intense blue color around the colonies (Table 1). Of these, 23 and 6 isolates were isolated from rubbish dump site and palm oil mill, respectively. Further confirmation test was performed quantitatively on the 29 putative lipase producers that gave positive results on triolein agar plates. Various basal media M1, M3, TYEM and BM-1 were tested for lipase production during the selection. Low lipase activities ranging from 0 ~ 0.044 U/ml were detected colorimetrically in culture supernatant, except for isolate T1 T , which showed high lipase activity (0.150 U/ml) after 24 h incubation (Table 1).
On the basis of relatively higher lipase activity detected for isolate T1 T , the effect of temperature on the activity and stability of crude T1 lipase was further investigated. The effect of temperature on the lipase activity and stability was examined from 40 to 80°C. As shown in Fig. 1, the crude enzyme from isolate T1 T manifested its maximal activity at 70°C with olive oil as substrate. Crude T1 lipase was fairly stable up to 60°C for 30 min and gradually decreased upon prolonged temperature treatment. This is due to T1 lipase tended to lose its native conformation as a result of breaking of the intrinsic interaction above its stable range. This is a discrepancy between temperature activity (70°C) and stability (60°C), as the enzyme is tend to be protected by heat denaturation in the presence of olive oil. In addition, the enzyme thermostability is greatly influenced by the presence of water, because denaturation is linked to its conformational mobility in aqueous mixture [34]. The crude enzyme of isolate T1 T was fairly active at higher temperature as compared to other thermostable lipases from Bacillus spp. [26,30,33] which exhibited maximal activity at 60°C. High temperature activity and stability of enzyme offer great potentials in industrial applications, and hence attempts have been made to identify isolate T1 T .

Identification of isolate T1 T
During the characterization of organism isolated from palm oil mill effluent, strain T1 T was recovered on nutri-ent broth at 60°C. The growth condition for strain T1 T was 50-70°C and between pH 5 and 9 with the optimum growth temperature and pH of 65°C and pH 6.5, respectively, in nutrient broth. These met the criteria of thermophilic bacteria, which grew at temperatures above 50°C [35]. To verify the systematic position of this bacterium, a study of morphological and physiological characteristics, 16S rRNA analysis, cellular fatty acids analysis, DNA composition, DNA/DNA hybridization, RiboPrint analysis, lipase gene analysis and protein profile were undertaken.
The cellular morphology of isolate T1 T is rod-shaped, 0.8-1.0 μm width and 2.5-6.0 μm length, gram positive bac-teria. The terminal spore is oval/cylindrical in shape and swollen the sporangium. The DNA base composition of strain T1 T is around 52.6% mol G + C. The partial sequencing of the 16S rDNA shows 99.5% similarity to validly described Geobacillus kaustophilus (DSM 7263 T ) and Geobacillus thermoleovorans (DSM 5366 T ). The 16S rRNA sequence of strain T1 T is a continuous stretch of 1519 bp (AY166603). Construction of phylogenetic trees using the neighbour-joining method in determining the evolutionary relationship among a group of validly described closely related species is indicated in Fig. 2. A comparison of biochemical, morphological and physiological properties of strain T1 T with its closest phylogenetic neighbors is presented in Table 2. Strain T1 T can be distinguished from Geobacillus thermoleovorans (DSM 5366 T ) phenotypically by oxidase test, arabinose, mannitol, inositol, lactose and casein hydrolysis. However, strain T1 T differs from Geobacillus kaustophilus (DSM 7263 T ) by lysozyme test, arabinose, mannitol, ribose, adonitol, lactose, gelatin and casein tests. Since the sequencing result and physiological data did not allow strain T1 T to be identified with one of the above mentioned species, further analysis need to be carried out to verify its phylogenetic position.
Result of chemotaxonomic analyses is given in the species description. The fatty acid profile of strain T1 T is typical for the Bacillus rRNA-group 5 (thermophilic Bacillus strains).
Lipase activity (■) and stability (▲) of crude T1 lipase at var-ious temperature Figure 1 Lipase activity (■) and stability (▲) of crude T1 lipase at various temperature. Crude T1 lipase was assayed at various temperatures ranging from 50 to 80°C with olive oil emulsion (1:1, v/v) as substrate (pH 7.0). For the lipase stability test, crude T1 lipase was assayed after heat treatment at various temperatures for 30 min.

Isolates
Tri olei n agar (at 60°C ) Lipase activity (U/ml) Rubbish dump site: Note: (+), indicates formation of intense blue colour on triolein agar plate The major content of cellular fatty acids of strain T1 T is isofatty acids. Among them, iso-branched pentadecanoic acid (iso-C15), hexadecanoic acid (iso-C16) and heptadecanoic acid (iso-C17) making up 78.33% of the total fatty acids for strain T1 T . Strain T1 T can be differentiated from Geobacillus thermoleovorans DSM 5366 T based on percent composition of iso-fatty acids (iso-C15, iso-C16 and iso-C17 making up 62.1%). There was only 6.14% of iso-C16 for strain T1 T but 21% for Geobacillus thermoleovorans DSM 5366 T (Table 3).
Although FAME is not always a reliable tool for preliminary identification of bacilli, it could be used in combination with other identification methods [36]. Further analysis was also carried out by amplifying fulllength thermostable lipase gene using primers as described in materials and methods [38]. Fig. 3 showed amplified full-length lipase genes of strain T1 T and its type strains. The amplified lipase gene of strain T1 T was around 2 kb but 1.8 kb for its type strains Geobacillus kaustophilus DSM 7263 T and Geobacillus thermoleovorans DSM 5366 T . A stretch of about 220 bp insertion could be seen at downstream of the open reading frame of thermostable lipase gene (Fig. 4). In addition, the intracellular protein profiles were determined by SDS-PAGE. Samples (30 μg) were separated on 12% SDS-polyacrylamide gel and stained using Coomassie blue. Strain T1 T showed obvious different protein profile as compared to its type strains at region between 34 to 47 kDa (Fig. 5).

Description of Geobacillus zalihae sp. nov
Geobacillus zalihae (za.li'ha.e. N.L. gen. n. zalihae of Zaliha). The novel species is isolated from palm oil mill efflu- Kaustophilus DSM 7263 T ). All strains were negative for anaerobic growth, growth at 30°C, VP reaction, indole production and acid from L-rhamnose, sorbitol. All strains were positive for catalase, hydrolysis of starch, use of citrate and acid from D-fructose and D-glucose. ent in Selangor, Malaysia, with the type strain T1 T (DSM 18318 T ; NBRC 101842 T ). Cells are rod-shaped, 0.8-1.0 width and 2.5-6.0 length, gram positive bacteria. The terminal spores are oval/cylindrical and swollen the sporangium. Growth occurs at 50-70°C with an optimum temperature of 65°C. Growth at 65°C occurs between pH 5 and 9 with maximal growth at pH 6.5. The DNA base composition of strain T1 T was around 52.6% mol G + C.
The iso-fatty acids were in major amount according to cellular fatty acid profile in which iso-C15 (32.42%), and iso-C17 (39.77%) were in abundant (77.19%). Growth is aerobic and tolerant up to 2% NaCl. It can not perform anaerobic growth. It shows positive in catalase test but not oxidase test. It is able to hydrolyze starch but not gelatin and casein. Acids are produced from L-arabinose, D-lactose but not D-mannitol.
Phylogenetic position of Geobacillus zalihae T1 T with other validly described species of the genus Geobacillus . Escherichia coli were used as an out-group. Phylogenetic tree was inferred by using the neighbour-joining methods. The software package MEGA 3.1 was used for analysis.

Conclusion
As a consequence, the strain T1 T merits recognition as a member of a novel species through morphological and physiological studies, cellular fatty acids composition, DNA composition, DNA/DNA hybridization, RiboPrint analysis. Sizes of full-length lipase genes and protein profiles were additional evidences. The Geobacillus zalihae strain T1 T was deposited in DSMZ (DSM 18318 T ) and NITE (NBRC 101842 T ).

Bacterial isolation
Numerous enrichment cultures that were derived from samples of rubbish dump site and palm oil mill effluent were screened for their ability to degrade olive oil.  [21], to which 0.01% of trace elements solution containing 0.026% B, 0.05% Cu, 0.05% Mn, 0.006% Mo and 0.07% Zn were added [27]. The enriched cultures were further screened by using triolein agar plate. Triolein agar comprising of triolein (0.25%), bacteriological agar (1%), nutrient broth (0.8%) and Victoria Blue (0.01%) was adjusted to pH 7.0. The medium was homogenized for 5 minutes before sterilization. The sterilized triolein agar was poured into petri dishes. Isolates that showed positive results on the triolein agar were then tested for their lipase production in basal media.

Selection of thermostable lipase producer
Lipase production was determined aerobically at 60°C in 500 ml blue cap bottle containing 100 ml basal medium. The composition of basal mineral media used in this study was (g/L): BM1 (  [21]. The pH was adjusted to 7.0 and the medium was sterilized for 15 min at 121°C. Bacterial inoculum (3 ml) was then inoculated into 100 ml basal medium and incubated by shaking at 150 rpm, 60°C. The cell free supernatant was obtained by centrifugation at 10, 000 rpm, 4°C for 10 min prior to lipase assay.

Lipase assay
The lipase activity was assayed by colorimetry [41]. Culture filtrate (1 ml) was shaken with 2.5 ml of olive oil (70% oleate residues) emulsion (1:1, v/v) and 20 μl of 0.02 M CaCl 2 in a water bath shaker at an agitation rate of 200 rpm. The emulsion was prepared by mixing together an equal volume of olive oil (Bertoli, Italy) and 50 mM phosphate buffer (pH 7.0) with a magnetic stirrer for 10 min. The reaction mixture was shaken for 30 min at 50°C. The enzyme reaction in the emulsion system was stopped by adding 6 M HCl (1 ml) and isooctane (5 ml), followed by mixing using a vortex mixer for 30 s. The upper isooctane layer (4 ml) containing the fatty acid was transferred to a test tube for analysis. Copper reagent (1 ml) was added and again mixed with a vortex mixer for 30 s. The reagent was prepared by adjusting the solution of 5% (w/ v) copper (II) acetate-1-hydrate to pH 6.1 with pyridine. The absorbance of the upper layer was read at 715 nm. Lipase activity was measured by measuring the amount of free fatty acids released based on the standard curve of free fatty acid. One unit of lipase activity was defined as the amount of enzyme releasing 1 μmole of fatty acid per minute.

Characterization of crude lipase
The effect of temperature of the crude lipase was evaluated by assaying at temperatures ranging from 50 to 80°C. Crude enzyme (1 ml) was shaken with 2.5 ml of olive oil (70% oleate residues) emulsion (1:1, v/v) and 20 μl of 0.02 M CaCl 2 in a water bath shaker at an agitation rate of 200 rpm. The lipase activity was measured colorimetrically. Assessment of the thermostability of crude lipase was performed by measuring the residual activity after 30 min pre-incubation at various temperatures ranging from 50 to 80°C. The treated enzyme was immediately put in ice-bath for 10 min before measuring the residual activity at 50°C for 30 min.

Identification of strain T1 T
A study of morphological and 16S rRNA analysis was conducted in UPM. Its physiological characteristics, cellular fatty acids analysis, DNA composition, DNA/DNA hybridization and RiboPrint analysis was undertaken in Deutsche Sammlung Von Mikroorganismen (DSMZ), Germany.

Morphological and physiological study
For the morphological study, pure bacterial strain was streaked on nutrient agar plate and incubated for 24 h at 60°C prior to gram staining. It was then observed under a light microscope. Morphological and physiological char-acteristics were further determined in DSMZ (Germany). The physiological characteristics study included catalase and oxidase test, anaerobic growth, Voges-Proskauer test, growth at 30, 40 and 70°C, growth in medium at pH 5.7, 2% and 5% NaCl, lysozyme broth, fermentation of D-glucose, L-arabinose, D-xylose, D-mannitol, D-fructose, Minosit, D-ribose, D-cellobiose, L-rhamnose, sorbitol, Dgalactose, adonit, D-lactose, hydrolysis of starch, gelatin, casein and Tween 80, decomposition of tyrosine, use of citrate and propionate, nitrate reduction, indol production, phenylalanine deaminase and arginine dihydrolase test.

Cellular fatty acids analysis
Fatty acids were extracted and analysed following the instructions of Sherlock microbial identification system. The culture of Bacillus thermocatenulatus was used as control during cellular fatty acids analysis.

16S rDNA analysis
The product was purified using QIAquick gel extraction kit (Qiagen, Germany). The purified PCR product was cloned into TOPO TA PCR 2.1 cloning vector (Invitrogen, USA). The recombinant plasmid was extracted with QIAprep plasmid extraction kit (Qiagen, Germany) and was then sequenced using an ABI PRISM 377 DNA sequencer (Applied Biosystems, USA) as recommended by the manufacturers. The 16S rDNA sequence of Geobacillus zalihae T1 T was analyzed using software package MEGA 3.1 [42].

DNA base composition
The chromosomal DNA was isolated and purified according to the procedure of Cashion et al. [43]. The G + C contents were determined by using chromatography conditions adopted from Tamaoka and Komagata [44]. The DNA was hydrolysed and the resultant nucleotides were analysed by reverse-phase HPLC [45].
DNA/DNA hybridization DNA/DNA hybridization was carried out as described by De Ley et al. [46], with the modifications described by Huss et al. [47] and Escara & Hutton [48], using a model 2600 spectrophotometer equipped with a model 2527-R thermoprogrammer and plotter (Gilford Instrument Laboratories). Renaturation rates were compared with the TRANSFER.BAS program by Jahnke [49].

Ribotyping analysis
Standardized automated ribotyping is performed using the QualiconTM RiboPrinter system as described by Bruce [50]. The RiboPrinter system combines molecular processing steps for ribotyping in a stand-alone, automated instrument. Steps included cell lysis, digestion of chromosomal DNA with restriction enzyme EcoR1, separation of fragments by electrophoresis, transfer of DNA fragments to a nylon membrane, hybridization to a probe generated from the rrnB operon from E. coli, chemiluminescent detection of the probe to the fragments containing rrn operon sequences, image detection and computerized analysis of RiboPrint patterns.

Amplification of full-length thermostable lipase gene
Genomic DNA was extracted by using conventional method [51]. In order to amplify the full length sequence of thermostable lipase gene, a set of primers was designed based on the thermostable lipase gene sequence of pmol), and Taq DNA polymerase (2 U). The gene was amplified with a thermocycler (Gene Amp PCR system 2400, Perkin Elmer, Foster, CA) with the temperature program of predenaturation at 94°C for 4 min, 30 cycles PCR of 1 min denaturation at 94°C, 2 min annealing at 55°C and 2 min extension at 72°C. The final extension step at 72°C was 7 min and preservation was at 4°C. The amplified products were electrophoresed on 1.0% agarose gel (w/v) with 1 kb marker as the standard marker. The PCR products were purified and sequenced as described earlier.
The sequences alignment was generated using CLUSTALW and TEXSHADE in Biology Workbench 3.2 [52].
Protein pattern analysis SDS-PAGE was done on 12% running gels by using the method of Laemmli [53]. A broad range of protein standard (MBI Fermentas, Germany) was used as a molecular mass marker. Extracellular protein was concentrated with 5,000 MWCO cut-off vivaspin 15R (Vivascience, Germany). Protein samples (30 μg) were loaded for analysis.

Authors' contributions
TCL performed most of the experiments described in this paper, contributed to their design and analysis, and helped to draft the manuscript. RNZRAR, ABS and MB conceived of the study and experimental design, and prepared the manuscript. All authors read and approved the final manuscript.