OUP user menu

The degradation of n-hexadecane in soil by thermophilic geobacilli

Roger Marchant, Freddie H. Sharkey, Ibrahim M. Banat, Thahira J. Rahman, Amedea Perfumo
DOI: http://dx.doi.org/10.1111/j.1574-6941.2006.00061.x 44-54 First published online: 1 April 2006


Soil microcosms have been used to demonstrate the ability of indigenous soil thermophiles to degrade effectively a representative alkane (n-hexadecane). A fragment of the alkane mono-oxygenase gene (alkB) was amplified from thermophilic Geobacillus thermoleovorans strain T70 by PCR using degenerate primers. The amplicon demonstrated 96% sequence similarity with the alkB gene from Rhodococcus erythropolis. Critical controls ensured that the positive PCR signal detected was not a result of mesophilic soil organisms. A reverse transcription PCR (RT-PCR) assay was developed to determine if expression of the gene was inducible in the presence of an alkane or constitutively expressed in soil. In the presence of n-hexadecane, expression of the alkane mono-oxygenase gene was induced in pure cultures and soil samples and was dependent on temperature. No positive RT-PCR signal was detected at mesophilic growth temperatures either in pure cultures or in soil microcosms, whereas at 55°C positive RT-PCR signals were obtained for both pure cultures of T70 and soil samples. Many different amplicons of the alkB gene fragment were obtained from the soil used in the microcosms. Thirty cloned fragments yielded 27 different sequences showing 85–96% sequence similarity with the alkB sequence of T70. To establish that the amplified alkB gene sequences from soil were derived from thermophilic geobacilli, additional strains were isolated on a selective medium containing n-hexadecane as sole carbon source. The 16S rRNA gene sequences were determined to identify the 50 isolates obtained (G. thermoleovorans, 27; G. caldoxylosilyticus, 17; G. pallidus, 2; G. toebiii, 1; Geobacillus sp., 3) representing 18 different strains and alkB gene sequences determined and deposited with the European Bioinformatics Institute.

  • reverse transcription PCR
  • alkane mono-oxygenase
  • gene expression
  • thermophile
  • Geobacillus


Interest in thermophilic bacilli has been recently rekindled by the reclassification of a number of Bacillus species into the new genus Geobacillus (Nazina et al., 2001) coupled with the description of additional species (Nazina et al., 2001; Sung et al., 2002; Banat et al., 2004). Although many of the Geobacillus isolates have come from high-temperature environments, it has been pointed out that they are frequent in temperate-climate soil environments where they are apparently unable to grow due to the temperature restriction (Marchant et al., 2002a, b). It has also been established that an extensive diversity of geobacilli, many of which have the capability to degrade alkanes and hydrocarbons, exist in these soil environments (Rahman et al., 2004). What has so far been lacking has been any effort to determine what role, if any, these highly thermophilic bacteria may play in soil environments at temperatures below their apparent growth minimum. We have attempted to address some aspects of this problem in the current study through an investigation of alkane degradation in soil microcosms. This is a relatively uncommon microbial pathway, which we have selected simply as a representative activity of geobacilli and not necessarily one which is crucial to their survival and growth, since they can utilize a wide range of substrates.

Microorganisms capable of degrading alkanes are readily isolated from contaminated and non-hydrocarbon-contaminated sites; for some of these organisms the genetic characteristics of their alkane degradative systems have been investigated (van Beilen et al., 2003). The oxidation of inert alkanes to a primary alcohol involves a three-component alkane mono-oxygenase complex, comprising an integral membrane mono-oxygenase and two soluble proteins, rubredoxin and rubredoxin reductase (Kusunose et al., 1967; Rehm & Reiff, 1981). Smits. (1999) reported that the alkane mono-oxygenase genes from different genera appear to be quite divergent. The degradative pathways have only been characterized from a small number of gram-positive and gram-negative bacteria, including Pseudomonas putida (van Beilen et al., 2001), Acinetobacter (Ratajczak et al., 1998a,b) and Rhodococcus erythropolis (Whyte et al., 2002a). Research has focused primarily on the organization and expression patterns of the genes encoding these proteins, especially in Pseudomonas oleovorans and Acinetobacter species.

The genes encoding the alkB complex in P. oleovorans have been shown to be closely grouped into two operons located on the OCT plasmid (van Beilen et al., 1992). The alkBFGHJKL operon encodes two components of the alkB system and other enzymes involved in further metabolic steps (van Beilen et al., 1994). The second operon encodes rubredoxin reductase, alkT and alkS, which has been shown to regulate expression of the entire alkBFGHJKL operon (Kok et al., 1989; van Beilen et al., 1994). alkS activates the expression of the promoter PalkB, which results in transcription of most genes in the first operon (Yuste et al., 1998). By contrast, the genes encoding the alkB complex in Acinetobacter sp. APD1 appear to be arranged in a totally irregular fashion (Ratajczak et al., 1998b). Alkane degradation in Acinetobacter is highly dependent on the expression of at least five genes, with alkR known to regulate transcription of the entire gene cluster (Ratajczak et al., 1998a). In Rhodococcus strains Q15 and NRRI, at least four alkane mono-oxygenase gene homologues (alkB1–alkB4) have been identified (Whyte et al., 2002b). The alkB1 and alkB2 homologues were part of the alkB gene clusters, encoding two rubredoxins, a transcriptional regulator protein and rubredoxin reductase. Homologues alkB3 and alkB4 were found as separate genes and not part of the alkB gene clusters.

We have investigated the effect of temperature on degradation of hexadecane, as a representative alkane, in soil microcosms and then supplemented this study with an investigation of expression of the alkane mono-oxygenase gene (alkB) in thermophilic geobacilli, in pure culture and in soil environments, using reverse transcription PCR (RT-PCR). Despite the availability of various assays for detection and localization of gene expression, RT-PCR is generally considered the most sensitive method for detecting low copy numbers of mRNA (Sharkey et al., 2004a, b). Because of the short half-life of bacterial mRNAs (1–3 min) (Belasco & Higgins, 1988; Arraiano, 1993), RT-PCR provides the opportunity of obtaining an immediate indication of the expression status. We have also determined the level of species diversity of the thermophilic geobacilli in our soil system (Rahman et al., 2004).

Materials and methods

Soil sample collection, strains and growth media

Soil samples were collected from a site within Northern Ireland. Site ‘T’ was under established mixed coniferous and deciduous trees with no ground-cover plants (Irish grid reference C88 216). Samples were taken at a depth of 50 mm into the mineral layer of the soil. Meteorological records over 30 years, of soil temperature at 50 mm subsurface, taken from the University of Ulster meteorological station, have indicated that the maximum temperature of the soil within site T would never have exceeded 25°C. Site T was known to contain sufficient numbers of thermophilic organisms, and was the site from which G. themoleovorans T70 was isolated. Enrichment of soil samples was carried out by inoculating nutrient broth with 0.1 g of soil and incubating at the desired temperature for 24–48 h, after which broth samples were streaked on solid nutrient medium plates. Pure cultures of three G. thermoleovorans strains initially examined were grown overnight at 70°C on nutrient agar slopes. Only strain T70 (16S rRNA gene accession number AJ489328) was further analysed for alkB gene expression analysis. When determining if the alkB gene from pure cultures of T70 was induced or constitutively expressed, strain T70 was grown for up to 48 h on mineral salts medium containing 1%n-hexadecane as the sole carbon source. Soil microcosms were established in glass universals using glass wool as an interface between 1 mL n-hexadecane and 10 g of soil. The microcosms were then incubated for up to 14 days at room temperature, and 25, 30, 37 and 55°C. A further soil microcosm was also inserted in situ just below the surface soil layer in the natural environment for the same period of time. Control microcosms containing no alkane were also prepared using each of the above conditions.

Additional thermophilic, alkane-degrading geobacilli were isolated by enrichment at 60°C from soil samples using 1%n-hexadecane as the sole carbon source in mineral salts medium plus 0.5% yeast extract. Pure cultures were obtained from single colonies on streak plates and 50 isolates representing 18 different strains as identified by 16S rRNA gene sequences.

Soil microcosms

Soil for use in the microcosms was sieved through a 1-mm screen. One portion of the soil was sterilized at 121°C for 30 min and then dried overnight at 60°C. The water content of nonsterile soil was determined by weight loss after drying (26% weight in weight, w/w). The water content of all experimental microcosms was adjusted to 30% (w/w) by adding sterile water. Microcosms were set-up in sterile screw cap glass universal bottles containing 5 g of soil. n-Hexadecane (Sigma, Poole, UK) was added at 2% (v/w). In microcosms that were nutrient-supplemented, a nitrogen, potassium and phosphorous additive (NPK) solution was used containing (per litre) 10 g (NH4)2NO3 and 10 g KH2PO4. Microcosms were incubated at room temperature (18°C) or at 60°C and all treatments were carried out in duplicate. The composition of the microcosms were as follows: SS+HY: 5 g sterile soil + 100 μL hexadecane+1400 μL sterile water NSS+HY: 5 g nonsterile soil+100 μL hexadecane+1400 μL sterile water SS+HY+NPK: 5 g sterile soil+100 μL hexadecane+500 μL NPK+900 μL sterile water NSS+HY+NPK: 5 g nonsterile soil+100 μL hexadecane+500 μL NPK+900 μL sterile water

Duplicate sets of microcosms were assayed for hexadecane content, using gas liquid chromatography at 0, 5, 15, 30 and 40 days.

Alkane degradation

Estimation of hydrocarbon degradation in the soil microcosms was determined using FID gas liquid chromatography (Perkin Elmer model 8600, Boston, MA). Hexane (J. T. Baker Laboratories, Phillipsburg, NJ) was used to extract the hexadecane from the soil (2 : 1 ratio). Prior to opening the microcosms for extraction they were allowed to cool to room temperature to reduce loss by volatilization. The soil and hexane were mixed for 30 min and soil-free hexane extract was recovered by filtration through a Minisart 0.2-μm filter and used for analysis. The conditions were as follows: injection temperature 250°C, detector temperature 250°C, column temperature programmed at 50°C for 4 min then increased at the rate of 20°C min−1 to 330°C for 3 min. The column used was a ZB5 column (Phenomenex, Macclesfield, UK). Hexadecane peaks were identified by retention time and comparison with standards. Hexadecane concentrations in the microcosms were determined by peak area comparison using initial spiked soil-extracted values as 100%.

Bacterial enumeration

Total bacterial counts of mesophilic and thermophilic bacteria in the soil microcosms were carried out at 0 and 40 days using dilution plating onto nutrient agar with incubation at either room temperature or 60°C as appropriate.


The sequences from which degenerate primers were designed for the alkB gene in this study can be found in the EMBL database under accession numbers AF388179 (alkB3, Rhodococcus sp. Q15), AF388180 (alkB4, Rhodococcus sp. Q15), AJ301877 (alkB4, Rhodococcus erythropolis NRRLB-16531) and AJ009587 (alkB, Prauserella rugosa).

Molecular biology

DNA was extracted from pure cultures of T70 and all soil samples using the BIO101 FastDNA® Spin kit (Bio 101 Inc, CA) according to the manufacturer's instructions. Extracted DNA samples were then analysed by gel electrophoresis (100 V for 60 min) on a 1.5% agarose gel (Invitrogen, Paisley, UK) containing 10 mg mL−1 ethidium bromide (Sigma, Poole, UK). PCR reactions were carried out on a Gradient Thermocycler (Whatman Biometra, Niedessachsen, Germany) containing the THERM or alkB gene-specific primers. The PCR was carried out in a 100 μL volume containing 1 μL of template DNA, 2 mM MgCl2, 0.2 mM of each dNTP, 50 pM of each primer, and 2.5 units of Taq DNA polymerase (Invitrogen). Each PCR cycle consisted of denaturation for 1 min at 94°C, annealing for 30 s at 53°C (THERM primers, see Table 1) or 58.5°C (alkB primers, see Table 1), and extension for 30 seconds at 72°C for a total of 38 cycles, followed by a final extension cycle at 72°C for 10 min. PCR products were sequenced using an ABI PRISM® 3100 Genetic analyser (Applied Biosystems, Foster City, CA). Prior to the actual sequencing reaction, the PCR products were purified using the Concert™ rapid purification system (Invitrogen). Cycle sequencing of the PCR products was carried out according to the manufacturer's recommendations, and precipitated with ethanol.

View this table:

PCR and reverse transcription primers

Primer nameForward primerReverse primer
Target gene 16S rRNARahman. (2004)(Universal reverse–1492R)
  • R, G or A; Y, C or T; S, G or C.

Extraction of RNA from pure cultures and from soil

RNA was isolated from T70 using a TNS extraction protocol as used by Sharkey. (2004a). The extraction buffer contained 1% triisopropylnaphthalene sulfonic acid (TNS) and 6%p-4-aminosalicyclic acid (PAS), 200 mM Tris/HCl, 25 mM EDTA and 250 mM NaCl (pH 7.8). Cells from an overnight culture of G. thermoleovorans T70 or 0.5 g of soil were suspended in a pre-chilled 1-mL lysozyme (50 mg mL−1) containing Tris EDTA buffer containing a specialized matrix consisting of 0.5 g of a mixture of ceramic and glass beads. The mixture was then vortexed vigorously for approximately 40 s, cooled briefly, and microcentrifuged at 4°C for 10 min at 14 000 g to enhance elimination of excessive cellular debris. The supernatant was removed and one volume of 1% TNS buffer added and allowed to settle on ice for 30 min. The samples were then microcentrifuged at 10 000 g for 10 min. Phenol chloroform isoamyl alcohol (Sigma) extractions were repeated at least twice until the upper aqueous phase was completely clear. Nucleic acids were precipitated with two volumes of 100% ethanol at −70°C for a minimum of 3 h followed by a final centrifugation at 16 000 g for 25 min. The pellet was allowed to air-dry while remaining on ice and resuspended in 30 μL of sterile RNAse-free water.

Quantification of extracted RNA was carried out using the RiboGreen® assay (Invitrogen) prepared in two formats, low and high range. The high-range assay (20 ng mL−1 to 1 μg mL−1 RNA final concentration in a total volume of 1 mL TE buffer (Tris-EDTA) containing 500 μL of 200-fold diluted RiboGreen®) proved to be most suitable and a standard curve was generated from Escherichia coli rRNA standards. The regression equation from the calibration (y=0.1464x+0.0616, R2=0.9996) was then used to determine experimental RNA concentrations.


Genomic DNA carried over during the RNA extraction procedure was digested with RNAse-free DNAse I (Roche). Ten Units of DNAse I was used to remove all traces of contaminating DNA in a 13.25-μL reaction also containing 50 mM MgCl2 and 10 μL of template RNA. The reaction mix was incubated for 30 min at 37°C. Effective DNAse I inactivation was achieved by adding 1 μL of 20 mM EDTA to each sample(s) and incubating at 37°C for 1 min followed by increasing the temperature to 65°C for a further 10 min.

Total RNA concentrations were determined using the RiboGreen® quantification assay according to the specifications of the manufacturer, with some minor modifications. Each RT reaction initially contained 200 ng of purified RNA, 50 pM of reverse primer, 2 μL of 10 mM dNTP mix, and made up to a volume of 12 μL with sterile DEPC (diethylpyrocarbonate)-treated water. The reaction mix was incubated at 65°C for 5 min followed by rapidly chilling on ice. Then 4 μL of reverse transcription buffer, 1 μL of 0.1 M dithiothreitol, 1 μL of RNAse Out Inhibitor, and 15 Units ThermoScript™ RT (Invitrogen) were added to each reaction mix. ThermoScript™ is engineered to have higher thermal stability (50–65°C) than other RT enzymes. The reaction was incubated for 50 min at the optimal annealing temperature for each primer set [THERM primer (53°C) and alkB primers (58.5°C)], followed by a further incubation at 85°C for 5 min to inactivate the activity of the ThermoScript. A critical control step where no ThermoScript was added was established to ensure only alkB-encoded mRNA was detected. Finally, 1 μL of RNAse H was added to each reaction mix and incubated at 37°C for 20 min. For the PCR step, 2 μL of cDNA was amplified using the same conditions as described for amplification of the alkB gene. All RT-PCR products were analysed on a 1.5% agarose gel containing 10 mg mL−1 ethidium bromide.

Cloning of alkB sequences from soil-extracted DNA and sequencing

Total DNA was extracted from a soil sample using the BIO101 FastDNA® Spin kit and then used as template with the alkB primer set as described above. The PCR amplicons were purified using the Concert™ rapid purification system (Invitrogen Life Technologies) prior to cloning using the TOPO TA Cloning® Kit Version Q (Invitrogen Life Technologies). Thirty clones were selected at random, the DNA was extracted from each clone and the cloned alkB gene was sequenced using M13 primers. Two clones produced identical sequences and one was therefore discarded; a further two clones failed to produce a recognizable alkB sequence and were discarded.

16S rRNA gene sequencing

The additional alkane-degrading Geobacillus isolates were identified from 16S rRNA gene sequences. The DNA extraction, PCR conditions and sequencing reactions were carried out as described by Marchant. (2002a).


Alkane degradation in soil microcosms

The time course for n-hexadecane degradation in the various microcosm treatments is shown in Fig. 1. The controls containing sterile soil show minimal loss of hexadecane over the entire 40-day experiment and the bacterial counts (Table 2) confirm that this small loss was not enacted by microorganisms. By contrast, the systems containing nonsterile soil showed progressive loss of alkane over the same time period, with nutrient supplementation producing enhanced degradation at both room temperature and 60°C. What is also clear is that degradation at 60°C is considerably greater than at room temperature, reaching more than 65%. In parallel with the degradation there was an increase in microbial numbers, thermophile numbers increasing by about 10-fold, whereas mesophile numbers only doubled. It should be noted, however, that the initial mesophile numbers were two orders of magnitude greater than the thermophiles. These results show that degradation of alkane is carried out at room temperature probably by mesophilic bacteria, but thermophiles are capable of achieving higher levels of degradation at appropriate temperatures.


Time course for n-hexadecane degradation in soil microcosms incubated at room temperature (RT, 18°C) and 60°C, using sterile soil (SS) and nonsterile soil (NSS), and with nitrogen, potassium and phosphorous (NPK) supplementation. Data points from duplicate samples with standard deviations.

View this table:

Viable cell counts for mesophilic and thermophilic bacteria taken from soil microcosms at 0 and 40 days. Results are means of five replicates

Treatments(CFU g of soil) at room temperature(CFU g of soil) at 60°C
At zero time
2NSS9.5 × 1051.2 × 104
3NSS+NPK7.5 × 1052.0 × 104
After 40 days
5NSS1.3 × 1068.5 × 104
6NSS+NPK1.5 × 1062.1 × 105
  • SS; sterile soil; NSS, nonsterile soil; NPK, nitrogen, potassium and phosphorous additive; ND, no colonies detected.

Geobacillus strains and soil samples

Four geobacillus strains (G. toebii F70 DSM 15390, G. thermoleovorans I80, T70 and T80 DSM 15905) obtained from soil samples in Northern Ireland and described in detail by Marchant. (2002a) were screened by PCR for the presence of the alkB gene. The numerical portion of each strain code refers to the initial enrichment temperature (Marchant et al., 2002a). Although these strains were obtained from enrichment cultures in the absence of any alkane, they have all been shown to have variable ability to degrade hydrocarbons (Marchant et al., 2002a). PCR products for the alkB gene were obtained from pure cultures of G. themoleovorans T70 only. This strain was therefore selected for further investigation. Soil samples known to contain this strain were also analysed.

Design of degenerate primers for PCR

Using sequence data derived from the European Molecular Biology Laboratory (EMBL) database, oligonucleotide primers specific for the alkB gene were designed. Various primer sets were designed initially using Primer Select software (DNAstar), to increase the chance of obtaining one or more positive PCR signals. Each primer set was designed around highly conserved regions of the alkB gene, from sequences of gram-positive organisms in the database. Alignment analysis of partial and complete alkB sequences was carried out using MegAlign software (DNAstar). Sequence alignments indicated that there were areas within each sequence that were highly conserved. Care was taken to design primers with minimal degeneracy, especially within the last three nucleotides at the 3′ end. Pure cultures of the four geobacillus strains were screened by PCR using the programme as described in the methods section for the presence of the alkB gene. A PCR product was obtained with G. themoleovorans T70. The fragment of the alkB gene amplified was of the expected size of 488bp using primers AF1 and AR1. DNA extracted from soil was then screened with primers AF1 and AR1. The PCR conditions and cycle numbers were the same as for the alkB amplification reaction from pure geobacillus cultures, except with the addition of Ultrapure Bovine Serum Albumin [(BSA), Ambion] to the reaction mix. Without BSA, no positive PCR signal was obtained from soil samples for the alkB gene. We believe that this may be due in part to the presence of PCR inhibitors in the soil samples. Owing to the specificity and sensitivity complications that can arise from detecting gene(s) directly from environmental samples by PCR, especially with the use of degenerate primers, it was necessary to add BSA to each PCR reaction mix. This proved successful, and a single positive signal(s) for the alkB gene fragment of the expected size was obtained by PCR (Fig. 2).


Detection of alkB gene fragments from pure cultures of geobacilli and soil samples. Lane 1, 100-bp DNA ladder. Lane 2, amplification of 270-bp fragment of the Geobacillus themoleovorans T70 16S rRNA gene using taxon-specific primer. Lane 3, detection of 488-bp alkB gene fragment from soil. Lane 4, no signal for mesophilic alkB primers. Lane 5, alkB gene fragment (488 bp) isolated from G. themoleovorans T70. Lane 6, negative DNA control.

Sequencing of the alkB gene fragment

The 488-bp alkB gene amplicons from both pure cultures and soil samples were sequenced using an ABI PRISM® 3100 genetic analyser. Sequencing analysis was carried out in both the 5′ and the 3′ directions using the alkB AF1- and AR1-specific primers, respectively. Both forward and reverse sequences were edited using Chromas 2.23 software. BLAST alignment analysis identified that both forward and reverse sequences demonstrated significant sequence similarity with database sequences for the alkB gene. The highest sequence similarity from pure cultures of T70 (alkB sequence accession number AJ781293) was 96% with the Rhodococcus erythropolis alkB gene for alkane-1-mono-oxygenase, strain 62–0. High sequence similarity was also seen with the alkB sequences from Rhodococcus sp. Q15 (95%) and an unidentified bacterium HXN 2000 (95%), but with smaller fragments of the alkB gene. A similar alkB gene sequence has also been obtained from another species of thermophilic soil bacterium, Geobacillus thermoglucosidasius TR2 (sequence accession number AJ781294).

Elimination of mesophile production of alkB in soil

It was essential to establish that the positive alkB PCR signal from soil samples was due to the presence of thermophilic geobacilli and not mesophilic organisms. One method that helped confirm this involved the use of a geobacillus taxon-specific 16S rRNA gene primer, which our research group have designed and analysed extensively (Rahman et al., 2004). This short primer of 15 bp was used as a positive control in all amplification reactions. This specific primer, for the 16S rRNA gene, designated THERM primer (for strain T70 position 1268–1283) was used with a universal reverse primer, 1492R, to produce a PCR product of 270 bp. A positive PCR signal using the THERM primer was obtained from all geobacilli pure cultures and soil samples examined (Fig. 2, lane 2). Additionally, enrichment analysis of soil samples was carried out on nutrient agar. Samples were incubated at 25°C for up to 2 days to allow growth of a mixture of mesophiles from the soil. Following DNA extraction from these mesophiles, no PCR signals for both the alkB gene and the 16S THERM primers further confirmed that alkB production did not originate from mesophilic organisms (Fig. 2, lane 4).


RNA was extracted from pure cultures of T70 and soil samples using a TNS–PAS-based technique (Sharkey et al., 2004a). The RNA extraction buffer containing TNS and PAS has been shown to be superior in terms of isolating mRNA to be used as a template in cDNA synthesis for RT-PCR (Bowler et al., 1999). The lysis buffer was used in conjunction with ceramic and glass beads for complete lysis. The integrity of the extraction procedure was clearly visible from the presence of sharp undegraded 23S (approximately 1-kbp mark), 16S (approximately 600-bp mark) and 5S (approximately 100-bp mark) rRNA bands during gel electrophoresis (Fig. 3). The relative intensity of each ribosomal band was taken as an estimate of the overall quality of the extraction procedure. The TNS–PAS method was initially compared with two commercially available kits. The TNS–PAS method proved superior in terms of RNA quality and quantity, as verified with the RiboGreen® quantification assay data. The RNA yields produced using the TNS–PAS method were greater than eightfold in comparison with one commonly used kit. We were unable to extract fully intact RNA using any method when G. thermoleovorans was grown above 55°C. RNA samples extracted above 55°C were extensively degraded. All DNAse I-treated RNA samples were quantified using the RiboGreen® quantification probe. Using pure cultures of G. thermoleovorans T70, we were able to determine that the alkB gene was induced in the presence of 1%n-hexadecane by RT-PCR analysis (Fig. 4). No RT-PCR signal was detected from strain T70 for the alkB gene when grown on nutrient broth slants containing no n-hexadecane. Soil microcosms containing the same concentration of n-hexadecane were then analysed to determine if the alkB gene was induced in the soil. Control microcosms containing no hydrocarbon were essential to establish that alkB gene expression was entirely due to the presence of the hydrocarbon, and not as a result of temperature induction. Positive RT-PCR signals for the alkB gene fragment were obtained from pure cultures of T70 and soil samples when incubated at 55°C. No positive RT-PCR signals were obtained in the absence of n-hexadecane, or indeed at temperatures below 55°C, which confirmed that there was no contribution from mesophiles to the RT-PCR signals.


Isolation of RNA from Geobacillus thermoleovorans T70 and soil samples. Lane 1, 100-bp DNA ladder. Lane 2, RNA isolated from pure culture of G. themoleovorans T70 grown at 50°C. Lane 3, RNA isolated from soil site T.


Detection of alkB gene expression in Geobacillus thermoleovorans T70 using reverse transcription PCR (RT-PCR). Lane 1, 100-bp DNA ladder. Lane 2, detection of alkB gene expression from pure cultures of T70 at 55°C in the presence of n-hexadecane. Lane 3, no product detected from pure cultures of T70 at 55°C in the absence of n-hexadecane. Lane 4, no product detected from soil at 25°C with n-hexadecane. Lane 5, no product detected from soil at 30°C with n-hexadecane. Lane 6, no product detected from soil at 37°C with n-hexadecane. Lane 7, detection of alkB gene expression from soil incubated at 55°C with n-hexadecane. Lane 8, no product detected from soil at 55°C without n-hexadecane. Lane 9, no product detected from soil at room temperature with n-hexadecane. Lane 10, no product detected from soil at room temperature without n-hexadecane. Lane 11, no product detected from soil in natural environment with n-hexadecane. Lane 12, negative RT-PCR control (no ThermoScript RT added).

Sequence diversity and identity of the alkB gene in soil

The inability to obtain sequence data for alkB amplicons for DNA extracted from the soil certainly stemmed from the mixed template originating from a range of different organisms. In order to establish that our PCR amplicons from soil were derived from thermophilic geobacilli and not mesophiles, the mixed PCR product was subjected to cloning and the alkB fragment from 30 clones was sequenced. Subsequently, new thermophilic Geobacillus pure cultures were isolated from the soil, using an alkane-enrichment method. The resulting isolates were identified using 16S rRNA gene sequences, as described previously (Rahman et al., 2004), and the sequence of the alkB gene determined. Clean alkB gene sequence data were obtained for one organism identified as Geobacillus thermoglucosidasius TR2 (16S rRNA gene accession number AJ781265, alkB sequence accession number AJ781294). The remaining isolates either produced no PCR product with the alkB primers or had different sequence gene copies of the same size, precluding the band stabbing of separated agarose gel DNA bands and repeat PCR of this template as was used to generate the sequence from G. thermoglucosidasius. The clone alkB sequences, the alkB sequence from the newly isolated G. thermoglucosidasius TR2, G. thermoleovorans T70 and selected alkB sequences for other organisms taken from the sequence database were used to construct a phylogram using the neighbour-joining method (Saitou & Nei, 1987) (Fig. 5).


Neighbour-joining tree of the alkB gene sequence from selected Rhodococcus species, Geobacillus thermoleovorans T70, Geobacillus thermoglucosidasius TR2 and 27 cloned sequences from soil. The range of sequence homology between the cloned thermophilic alkB gene from soil and the closest authentic database sequence was 65–82%.


Although there has been interest in the identification and diversity of members of the genus Geobacillus, and indeed the recent publication of a complete genome sequence for G. kaustophilus (an alkane degrader, but apparently lacking the alkB gene) (Takami et al., 2004), little effort has so far been devoted to establishing the role of these organisms in different environments. As part of a continuing study of these organisms, we have selected the hydrocarbon-degrading activity of geobacilli as a tool to examine their activity in temperate climate soil environments. The experiments using microcosms clearly show that some alkane degradation takes place at temperatures below the apparent minimum for thermophile growth, which is probably brought about by mesophilic bacteria, but which could be due to the growth of thermophiles at a slow rate in the special environment of the soil. By contrast, the degradation taking place at 60°C can only be attributed to the thermophiles and this conclusion is supported by the 10-fold increase in numbers of thermophiles which takes place under these conditions. Many of the hydrocarbon-oxidizing geobacilli have been isolated from formation waters of oil fields in Russia, Kazakhstan and China (Nazina et al., 2001), where the ambient temperature would allow their growth. In other environments where hydrocarbons are absent the geobacilli have a wide range of other substrates they can utilize.

Despite the detection and isolation of alkB genes from a wide variety of organisms with alkane-degrading capabilities, there have been no reports of detection of such genes from thermophilic geobacilli. Numerous strains of G. thermoleovorans have been shown to have strong alkane-degrading capabilities in our laboratory, and a number give a positive PCR signal with the alkB primers. However, only strain T70 was used for the purposes of this study.

A critical aspect of our study involved the elimination of mesophilic organisms present in the soil as sources of the alkB gene. This was particularly important considering that the design of the degenerate primer was based extensively on mesophilic Rhodococcus sequences for the alkB gene. Enrichment analysis of soil samples at mesophilic growth temperatures was carried out for up to 48 h, and screened by PCR for the alkB gene. This procedure was repeated several times and no positive PCR signal being detected on any occasion. Soil samples that were enriched at mesophilic growth temperatures were even investigated with a taxon-specific geobacillus primer set designed in our laboratory (Rahman et al., 2004). No positive PCR signal was detected using the 16S primer (THERM primer), confirming the probable linkage between the alkB signal and the presence of thermophilic geobacilli in the soil and establishing that the mesophilic enrichments contained no thermophiles. A clear, sharp, distinct band of 270 bp for the THERM primer was consistently obtained from all the pure geobacilli culture stocks in our laboratory. These findings in addition to positive PCR signals for the soil samples examined in the current study further confirmed that the alkB gene fragment detected was from thermophiles only. The fact that an RT-PCR product was obtained from soil microcosms incubated at 55°C for 2 weeks with an alkane and not at lower temperatures also indicates that mesophiles are not responsible for the positive alkB signal.

The RT-PCR experiment was designed to indicate if alkB gene expression in these thermophilic organisms was induced or constitutively expressed. Inducibility of the alkB gene was examined using n-hexadecane as the sole alkane source. Interestingly, it was found that in pure cultures of G. thermoleovorans T70, expression of the alkB gene was induced when grown at 55°C in the presence of n-hexadecane (Fig. 4, lane 2). Because of the well-documented difficulties encountered in extracting RNA from organisms grown at thermophilic temperatures (Luo & Stevens, 1997; Sharkey et al., 2004a, b), inducibility could not be examined at temperatures above 55°C. Soil microcosms containing n-hexadecane were incubated at 25, 30, 37 and 55°C together with a control microcosm containing no alkane at 55°C. This control was important, as it would indicate if the alkB gene could be induced at thermophilic growth temperatures in the absence of an alkane. Additional microcosms containing n-hexadecane and controls where no n-hexadecane was added were kept at room temperature and ambient temperature (data at ambient temperature with no n-hexadecane, i.e. no RT-PCR signal, are not shown). All microcosms were incubated for 1 week. The results indicated that alkB gene expression from soil samples was only induced in the presence of n-hexadecane (Fig. 4, lane 7) when incubated at 55°C. No positive RT-PCR signals were detected from any of the other microcosms, and the negative RT-PCR control used (Fig. 3, lanes 2–6 and 8–12). The whole procedure was repeated twice with the same outcome. A further microcosm was kept at ambient temperature in the presence of n-hexadecane for an additional 7 days, and examined by RT-PCR analysis to investigate if time-scale was related to induction of the alkB gene in the natural environment. However, no RT-PCR signal was detected (data not shown).

To eliminate further the possibility that mesophilic organisms were contributing to the soil RT-PCR signal and to link the observed signal to thermophilic geobacilli, the mixed alkB amplicons from the soil were cloned and sequenced. The neighbour-joining tree (Fig. 5) illustrates the wide sequence variability, with no clone sequence being identical to that of T70 despite the fact that T70 was isolated from the same soil. Only two of the clones had identical sequences. A BLAST search for each of the clone sequences gave results with either Rhodococcus sequences or unidentified organism sequences, but the sequence similarity for the best matches ranged from 65 to 82%. Interestingly the G. thermoleovorans T70 and the G. thermoglucosidasius TR2 sequences are the closest to the Rhodococcus sequences. Thus, of the 27 clone sequences used, none shows a high sequence similarity with Rhodococcus or any other mesophilic organism in the database, but they show higher sequence similarity with G. thermoleovorans T70 (85–96%). There is therefore clear evidence that many of the clone sequences have probably come from thermophilic geobacilli while conversely there is no evidence to suggest that any of the clone sequences have come from mesophilic bacteria.

The results show that the alkB gene in thermophilic geobacilli is induced by n-alkanes, rather than being constitutively expressed in the natural environment. These findings are consistent with previous studies, which have also shown that alkB gene expression is induced by the availability of specific or multiple alkanes. Transcription of the alkB gene of Burkholderia cepacia RR10 has been shown to be induced by n-alkanes of chain lengths 12–30 (Marin et al., 2001). It was also shown using transcriptional fusion assays that the alkB gene from Acintetobacter sp. strain ADP1 was induced to varying extents by a range of n-alkanes (Ratajczak et al., 1998b). In addition, our results indicate that although expression of the alkB gene is induced by alkanes, this induction is also highly influenced by temperature. This correlation confirms that transcription of the alkB gene in thermophilic geobacilli is highly regulated by both factors in synergy. It also further confirms that these organisms must have very limited metabolic activities in temperate-climate soil environments, but their role in the general ecology of such environments remains unclear and should be investigated further.


This work was supported by a grant number F/00430/A from the Leverhulme Trust and the C.E.C. EU Structural Funds, Building Sustainable Prosperity, Measure 5.1 ‘Sustainable Management of the Environment and Promotion of the Natural and Built Heritage (BSP7473), administered by the Environment and Heritage Service, Northern Ireland.


  • Editor: Jim Prosser


View Abstract