Abstract

The marine oil-degrading bacterium Alcanivorax borkumensis SK2 has attracted significant interest due to its hydrocarbonoclastic lifestyle, its alkane-centered metabolism, and for playing an important ecological role in cleaning up marine oil spills. In this study, we used microarray technology to characterize the transcriptional responses of A. borkumensis to n-hexadecane exposure as opposed to pyruvate, which led to the identification of a total of 220 differentially expressed genes, with 109 genes being upregulated and 111 genes being downregulated. Among the genes upregulated on alkanes are systems predicted to be involved in the terminal oxidation of alkanes, biofilm formation, signal transduction, and regulation.

Introduction

Marine oil-degrading bacteria play an essential role in degrading crude oil and thus in cleaning up marine oil spills (Yakimov, 2007). Alcanivorax borkumensis has become a paradigm of marine ‘hydrocarbonoclastic’ bacteria, as it exclusively grows on alkanes and plays a predominant ecological role in oil-degrading consortia that form following marine oil spills (McKew, 2007; Gertler, 2009). Alcanivorax borkumensis SK2 metabolizes a wide range of alkanes, such as linear alkanes, cyclo-alkanes, and isoprenoids (Dutta & Harayama, 2001; McKew, 2007). Given its important ecological role in the removal of oil spills and with the availability of its full genome sequence (Schneiker, 2006), Alcanivorax may now serve as a model organism to understand bacterial alkane metabolism.

Previous studies of bacterial alkane metabolism have been mostly focused on studying the terminal degradation of alkanes by Pseudomonas putida GPO1, a bacterium that uses linear alkanes as a single carbon and energy source, where a monooxygenase encoded by the alkB1 gene plays a central role in the terminal oxidation of alkanes (van Beilen, 1994). Other alkane-degrading bacteria use for this initial oxidation step enzymatic systems other than AlkB (for reviews, see van Beilen & Funhoff, 2007; Wentzel, 2007). Our genome-wide study of alkane utilization by A. borkumensis using a proteomics approach has revealed several alternative systems for terminal oxidation of alkanes by this bacterium, as well as major rearrangements of its central carbon metabolism (Sabirova, 2006). However, a number of specific questions intrinsically linked to alkane utilization by this organism, for example how alkanes enter the cell and which transport systems may be involved, how the cells physically interact with the hydrophobic substrate, whether and how they attach to it, and which molecular mechanisms allow the cells to protect themselves against the toxic effect of alkanes, are left unanswered. Finally, the regulatory implications of alkane degradation on the overall cellular activity could not be comprehensively studied using the proteomic approach. To obtain a still more comprehensive picture of alkane utilization, and in particular to be able to look more closely into some of the aforementioned issues, we have now used microarray technology to compare the transcriptional profile of SK2 grown on n-hexadecane, as a model alkane, as compared with pyruvate, one of the few non-alkane substrates A. borkumensis is able to use.

Materials and methods

Bacterial strains and growth conditions

Alcanivorax borkumensis SK2 was used for all experiments. Alcanivorax borkumensis was grown until the late-exponential stage of growth as described earlier (Sabirova, 2006). Bacteria from alkane- and pyruvate-grown cultures were centrifuged for 10min at 8000g, and the cell pellets were immediately frozen in liquid nitrogen and conserved at −80°C until RNA was isolated.

Oligonucleotide design, printing, and layout of the A. borkumensis microarray

The Abo3kOLI microarray used in this study is based on the sequenced genome of A. borkumensis (Schneiker, 2006). The array contains 2924 50mer to 70mer oligonucleotides representing predicted protein-encoding genes. In addition, the array contains 15 stringency controls of the genes gap, rpsA, rpsO, rpsP, and rpmI (70%, 80%, and 90% identity to the native sequence), 12 alien DNA oligonucleotides, and five spiking control oligonucleotides. Oligonucleotides were designed using oligodesigner software (Bioinformatics Resource Facility, CeBiTec, Bielefeld University). All oligonucleotide probes were printed in four replicates. Microarrays were produced and processed as described previously (Brune, 2006). Oligonucleotides (40μM) in 1.5M betaine, 3 × SSC (1 × SSC is 0.15M sodium chloride, 0.015M sodium citrate) were printed onto Nexterion Slide E (Schott AG, Mainz, Germany) using the MicroGrid II 610 spotter (BioRobotics, Cambridge, UK) equipped with 48 SMP3 stealth pins (TeleChem International, Sunnyvale, CA). DNA was cross-linked to the surface by incubation of the slides for 2h at 85°C. Microarray data have been submitted to ArrayExpress under accession number A-MEXP-1990.

RNA isolation and synthesis of labeled cDNA

Total RNA was purified using the RNeasy Mini Kit (Qiagen, Hilden, Germany). Cells were disrupted in RLT buffer provided with the Kit in Fast Protein tubes (Qbiogene, Carlsbad, CA) using the Ribolyser (Hybaid, Heidelberg, Germany) (30s, level 6.5) before spin column purification according to the RNeasy Mini Kit RNA purification protocol.

Fluorescent-labeled amplified RNA was prepared using the MessageAmp II-Bacteria RNA Amplification Kit (Applied Biosystems, Darmstadt, Germany). Starting from 500ng total RNA, cDNA carrying a terminal T7 promoter was synthesized. Subsequent in vitro transcription resulted in aminoallyl-modified RNA that was labeled with Cy3- or Cy5-N-hydroxysuccinimidyl ester dyes (GE Healthcare, Little Chalfont, UK). Uncoupled dye was removed applying the RNeasy MinElute Kit (Qiagen).

Microarray hybridization and image acquisition

Processing of microarrays before hybridization included the following washes: once in 0.1% Triton-X100 (5min, 20°C); twice in 0.032% (w/v) HCl (2min, 20°C); once in 0.1M KCl (10min, 20°C); once in H2O (1min, 20°C); once in 0.064% (w/v) HCl, 1 × Nexterion blocking solution (Schott AG) (15min, 50°C); and once in H2O (1min, 20°C). Microarrays were dried by centrifugation (3min, 185g, 20°C).

Hybridization was performed in an EasyHyb hybridization solution (Roche, Mannheim, Germany) supplemented with sonicated salmon sperm DNA at 50μgmL−1 in a final volume of 100μL for 90min at 45°C using the HS 4800 hybridization station (Tecan Trading AG, Switzerland). Before application to the microarrays, labeled samples were denatured for 5min at 65°C.

After hybridization microarrays were washed once in 2 × SSC, 0.2% sodium dodecyl sulfate (SDS) (w/v) (5min, 42°C), twice in 0.2 × SSC, 0.1% SDS (w/v) (1min, 21°C), twice in 0.2 × SSC (1min, 21°C), and once in 0.05 × SSC (1min, 21°C). Following the washes, slides were dried by centrifugation (3min, 185g, 20°C) and scanned with a pixel size of 10μm using the LS Reloaded microarray scanner (Tecan Trading AG). Four independent biological replicates including a dye swap were processed for each comparison.

Transcriptome data analysis

The mean signal and the mean background intensities were obtained for each spot of the microarray images using the imagene software 6.0 software (Biodiscovery Inc., Los Angeles) for spot detection, image segmentation, and signal quantification.

Spots were flagged as ‘empty’ if R≤0.5 in both channels, where R=(signal mean−background mean)/background SD. The remaining spots were considered for further analysis. After subtractions of the local background intensities from the signal intensities and the introduction of a floor value of 20, the log2 value of the ratio of intensities was calculated for each spot using the formula Mi=log2(Ri/Gi). Ri=Ich1(i)−Bgch1(i) and Gi=Ich2(i)−Bgch2(i), where Ich1(i) or Ich2(i) is the intensity of a spot in channel 1 or channel 2 and Bgch1(i) or Bgch2(i) is the background intensity of a spot in channel 1 or channel 2, respectively. The mean intensity was calculated for each spot, Ai=log2(RiGi)0.5 (Dudoit, 2002). A normalization method based on local regression was applied according to Yang (2002), Mi=log2(Ri/Gi)→log2(Ri/Gi)−c(A)=log2(Ri/[kj(A)Gi]), where c(A) is the LOWESS (locally weighted scatter plot smoothing) fit to the MA plot.

Significant up- or downregulation of genes was identified by t statistics (Dudoit, 2002). Genes were accounted as differentially expressed if P≤0.05 and M≥1.00 or ≤−1.00.

Normalization and t statistics were carried out using the emma 2.2 microarray data analysis software developed at the Bioinformatics Resource Facility, Center for Biotechnology, Bielefeld University (Dondrup, 2003).

Electron microscopy

For scanning electron microscopy (SEM), cells were grown on Permanox slides in ONR7a with either 1.5% hexadecane or 2% pyruvate as the carbon/energy source. SEM was carried out as described by Lünsdorf (2001).

Results and discussion

The microarray experiments were performed with the exponentially grown cells grown on either hexadecane or pyruvate (as control conditions), and led to the identification of a total of 220 differentially expressed genes, with 109 genes being upregulated and 111 genes being downregulated. Differentially expressed genes could be grouped into 15 functional categories, according to designated metabolic functions of the corresponding gene products. Both upregulated and downregulated genes were found in most groups, with the exception of those genes grouped under ‘alkane oxidation’, ‘stress’, and ‘iron uptake’, whose functions were exclusively induced in the presence of alkanes. ‘Nitrogen assimilation’ genes were all found to be expressed on pyruvate only, as were a number of other genes known to enable the cells to assimilate essential macroelements other than N, namely phosphorus and sulfur from less favorable sources. This effect may at least partially be attributed to higher cell densities present in pyruvate cultures, leading to some scarcity of these macroelements in the pyruvate, but not yet in the alkane-grown cultures. In the following, we therefore focus primarily on the functions that were found to be upregulated on alkanes, and thus can most clearly be attributed to A. borkumensis responses to growth on alkanes.

Terminal oxidation of alkanes

The presence of an enzymatic system mediating the terminal oxidation of alkanes distinguishes an alkane-degrader from a non-alkane-degrading organism. Our earlier proteomic study has already revealed the presence of several alternative ways for the terminal oxidation of alkanes by A. borkumensis (Sabirova, 2006). In accordance with the proteomic data, here, we find alkane monooxygenase alkB1 (ABO_2707, Table 1) to be upregulated on hexadecane. Moreover, a second alkane monooxygenase alkB2 (ABO_0122, Table 1) was also found to be upregulated, which corresponds to data using earlier reverse transcriptase-PCR (Schneiker, 2006). In addition, two flavin-binding monooxygenases were found to be upregulated during growth on alkanes indicative of two novel pathways likely to be involved in alkane degradation by A. borkumensis (ABO_0282, ABO_1097, Table 1). Moreover, we detected the up-expression of two genes similar to the ones involved in the degradation of halogenated alkanes in other bacteria, namely haloacid dehalogenase-like hydrolase dhlA (ABO_1537, Table 1) and haloalkane dehalogenase dhmA (ABO_2415, Table 1). If the first enzyme is known to convert haloalkanes to corresponding alcohols and halides, the second one catalyzes hydrolytic cleavage of carbon-halogen bonds in halogenated aliphatic compounds, leading to the formation of primary alcohols, halide ions, and protons. Alkane-induced coexpression of these enzymes mediating the breakdown of haloalkanes, alongside the induction of enzymes degrading aliphatic alkanes, signifies unspecific upregulation of expression, probably reflecting the presence of halogenated alkanes in sea water. Additionally, we found alkane-induced expression of aldehyde reductase (ABO_2414, Table 1). This gene is predicted to be involved in the metabolic activation of polycyclic aromatic hydrocarbons (PAHs), as shown recently for human aldehyde reductase AKR1A1 (Palackal, 2001). However, as yet, A. borkumensis has not been shown to either degrade or transform PAHs, and thus requires further experimentation to explore what coexpression of this gene alongside those mediating the degradation of aliphatic alkanes may signify for the degradation of alkanes or petroleum. These data allow us to update the list of enzymatic systems shown before by our proteomic study to be potentially involved in the initial terminal oxidation of alkanes by A. borkumensis (Figs 1 and 2).

1

Differentially expressed genes with known functions

Gene number Gene function or functional category ER SD Gene number Gene function or functional category ER SD 
Terminal oxidation of alkanes 
ABO_0122 Alkane-2-monooxygenase, AlkB2 +12.86 ± 0.40 ABO_0145 Ferredoxin – NADP+reductase, Frp +10.14 ± 0.57 
ABO_0282 Monooxygenase, flavin-binding family +15.32 ± 0.30 ABO_1097 Monooxygenase, flavin-binding family +7.81 ± 0.72 
ABO_1537 Haloacid dehalogenase-like hydrolase, DhlB +18.75 ± 0.28 ABO_1717 Flavodoxin reductase +8.51 ± 0.55 
ABO_2107 Monooxygenase, putative +4.77 ± 0.88 ABO_2414 Aldehyde reductase, AKR1A1 +14.02 ± 0.41 
ABO_2415 Haloalkane dehalogenase, DhmA +12.52 ± 0.40 ABO_2483 Alcohol dehydrogenase +6.39 ± 0.64 
ABO_2707 Alkane 1-monooxygenase, AlkB1 +6.67 ± 0.72 ABO_2845 Ferredoxin, 2Fe-2S +6.48 ± 0.67 
Energy and carbon metabolism 
ABO_0543 Respiratory nitrate reductase 1 gamma chain, NarI +3.66 ± 1.79 ABO_1753 3-dimethylubiquinone-9 3-O-methyltransferase, UbiG +3.34 ± 1.26 
ABO_2036 Cytochrome c oxidase, subunit I, CyoB +5.75 ± 0.79 ABO_1238 Enoyl-CoA hydratase/isomerase, putative +4.79 ± 0.89 
ABO_1652 Fatty oxidation complex, FadB2 +11.01 ± 0.46 ABO_1653 3-oxoacyl-CoA thiolase, FadA2 +9.20 ± 0.60 
ABO_1701 Acyl-CoA dehydrogenase, C-terminal domain, AcdA +5.83 ± 0.87 ABO_2102 Acyl-CoA dehydrogenase, putative +9.03 ± 0.59 
ABO_1445 Sugar aldolase, putative +5.11 ± 0.86 ABO_2741 Isocitrate lyase +23.17 ± 0.20 
ABO_0211 Phosphoglucomutase/phosphomannomutase −4.37 ± 0.93 ABO_0426 Glycerophosphoryl diester phosphodiesterase precursor −5.95 ± 0.79 
ABO_1042 Glycerophosphoryl diester phosphodiesterase, putative −8.07 ± 0.52 ABO_2613 Phosphoglycerate kinase, pgk −4.38 ± 0.90 
ABO_2701 2′,3′-cyclic-nucleotide 2′-phosphodiesterase, CpdB −3.67 ± 1.15 ABO_0840 Adenylate kinase, Adk −8.47 ± 0.50 
ABO_0885 Adenylate cyclase −6.40 ± 0.69     
Stress-related genes 
ABO_2633 DNA-binding stress protein +3.95 ± 1.22 ABO_0314 Heat shock protein, DnaK +11.32 ± 0.39 
ABO_0315 Heat shock protein, DnaJ +14.67 ± 0.34 ABO_0633 Chaperonin, 10 kDa, GroES +9.17 ± 0.51 
ABO_0634 GroEL chaperonin, Cpn60 +17.84 ± 0.30 ABO_1180 Heat shock protein, HtpX +11.67 ± 0.54 
ABO_1489 Heat shock protein, HtpG +13.37 ± 0.42 ABO_1777 Heat-shock protein, IbpA +17.45 ± 0.27 
ABO_2244 Heat shock protein, HslV +11.67 ± 0.49 ABO_2245 Heat shock protein, HslU +9.72 ± 0.51 
ABO_0095 Lactoylglutathione lyase, GloA +10.22 ± 0.56 ABO_0649 Superoxide dismutase, SodB +9.99 ± 0.40 
ABO_1531 Protein-methionine-S-oxide reductase, MsrB +7.05 ± 0.70 ABO_2703 Peptide methionine sulfoxide reductase, MsrA +3.81 ± 1.13 
ABO_0758 TolB protein +6.86 ± 0.65     
Transport systems 
ABO_2119 Flavohemoprotein, HmpA +13.29 ± 0.41 ABO_2370 Fe3+-hydroxamate ABC transporter protein +6.60 ± 0.61 
ABO_2487 Ferripyoverdine receptor precursor, FpvB +6.97 ± 1.14 ABO_1360 Copper-translocating P-type ATPase +13.64 ± 0.34 
ABO_1364 Copper-binding protein, CopC +9.83 ± 0.57 ABO_1411 OmpA family protein +11.81 ± 0.46 
ABO_0618 Sodium/alanine symporter, Dag A −6.71 ± 0.61 ABO_0851 Nitrate transport ATP-binding protein, NtrC −3.40 ± 2.93 
ABO_0852 Nitrate ABC transporter permease protein, NtrB −3.93 ± 1.07 ABO_1668 Ion transport protein −4.61 ± 0.93 
ABO_1888 Sulfonate ABC transporter protein −5.45 ± 0.75 ABO_2509 Branched-chain amino acid ABC transporter −3.88 ± 1.23 
ABO_2512 Branched-chain amino acid ABC transporter, ATP-binding protein, putative −6.40 ± 0.89 ABO_2513 Amino acid ABC transporter, ATP-binding protein −5.35 ± 0.92 
ABO_2544 Small multidrug resistance protein −9.19 ± 0.48 ABO_2684 Phosphate ABC transporter permease protein −6.81 ± 0.85 
Cell wall metabolism 
ABO_0601 Cell division protein, FtsQ +3.84 ± 1.28 ABO_0603 Cell division protein, FtsZ +10.02 ± 0.45 
ABO_0120 Amidase, AmiC +6.19 ± 0.73 ABO_1957 Membrane-bound lytic murein transglycosylase B +4.60 ± 0.90 
ABO_0260 Effector of murein hydrolase, putative −4.00 ± 1.12 ABO_1221 Membrane-bound lytic murein transglycosylase D −7.82 ± 0.58 
ABO_1425 Soluble lytic transglycosylase, putative −11.06 ± 0.39 ABO_2215 Heptosyltransferase II, putative −11.43 ± 0.54 
Biofilm, formation of alginate 
ABO_0395 Mannose-6-phosphate isomerase, AlgA +12.31 ± 0.48 ABO_0126 Type-1 fimbrial protein, A chain precursor +15.45 ± 0.39 
ABO_0584 Phosphoheptose isomerase, GmhA +6.63 ± 0.62 ABO_0463 Pilus biogenesis protein, PilN −6.59 ± 0.62 
ABO_0467 Pilin biogenesis-related protein −10.02 ± 0.40 ABO_0613 Type IV pilus assembly protein, TapC −10.74 ± 0.53 
ABO_0614 IV fimbrial assembly protein, PilB −6.14 ± 0.74 ABO_2670 IV pili twitching motility protein, PilT −5.19 ± 0.78 
Transcriptional regulators 
ABO_0031 Transcriptional regulatory protein, Qseb +7.52 ± 0.60 ABO_0121 Transcriptional regulator, GntR family +9.19 ± 0.48 
ABO_1680 DnaK suppressor protein homolog, DksA1 +4.37 ± 0.93 ABO_1708 Sensor histidine kinase +11.58 ± 0.38 
ABO_1709 DNA-binding response regulator +10.82 ± 0.38 ABO_2132 HD-GYP domain protein +3.79 ± 1.26 
ABO_0005 DNA-binding protein H-NS −5.36 ± 0.76 ABO_0104 Two-component hybrid chemotactic sensor and regulator, ChpA, putative −7.52 ± 0.94 
ABO_0167 Phosphate regulon sensor protein, PhoR −7.48 ± 0.64 ABO_0428 DNA-binding response regulator, LuxR family −11.76 ± 0.35 
ABO_0472 Two-component sensor histidine kinase, PilS −8.24 ± 0.49 ABO_0859 Nitrogen assimilation response regulatory protein −7.40 ± 0.63 
ABO_1745 Stringency control −6.78 ± 0.65 ABO_1840 Stringency control −11.76 ± 0.34 
ABO_1986 Two-component sensor histidine kinase protein, putative −7.13 ± 1.02 ABO_2018 Sensor histidine kinase/response regulator −10.05 ± 0.53 
ABO_2023 Catabolite gene activator, Crp −9.52 ± 0.52 ABO_2260 Sensory box histidine kinase, NtrB −6.55 ± 0.61 
ABO_2433 Conserved hypothetical signal transduction protein −7.14 ± 0.55 ABO_2508 Transcriptional regulator, GntR family −16.05 ± 0.35 
ABO_2519 Response regulator −11.50 ± 0.37 ABO_2520 Sensor histidine kinase −13.98 ± 0.29 
ABO_2530 Nitrogen regulatory protein P-II, GlnK2 −7.49 ± 0.56 ABO_2581 Transcription regulator family protein, AcrR −8.39 ± 0.51 
Gene number Gene function or functional category ER SD Gene number Gene function or functional category ER SD 
Terminal oxidation of alkanes 
ABO_0122 Alkane-2-monooxygenase, AlkB2 +12.86 ± 0.40 ABO_0145 Ferredoxin – NADP+reductase, Frp +10.14 ± 0.57 
ABO_0282 Monooxygenase, flavin-binding family +15.32 ± 0.30 ABO_1097 Monooxygenase, flavin-binding family +7.81 ± 0.72 
ABO_1537 Haloacid dehalogenase-like hydrolase, DhlB +18.75 ± 0.28 ABO_1717 Flavodoxin reductase +8.51 ± 0.55 
ABO_2107 Monooxygenase, putative +4.77 ± 0.88 ABO_2414 Aldehyde reductase, AKR1A1 +14.02 ± 0.41 
ABO_2415 Haloalkane dehalogenase, DhmA +12.52 ± 0.40 ABO_2483 Alcohol dehydrogenase +6.39 ± 0.64 
ABO_2707 Alkane 1-monooxygenase, AlkB1 +6.67 ± 0.72 ABO_2845 Ferredoxin, 2Fe-2S +6.48 ± 0.67 
Energy and carbon metabolism 
ABO_0543 Respiratory nitrate reductase 1 gamma chain, NarI +3.66 ± 1.79 ABO_1753 3-dimethylubiquinone-9 3-O-methyltransferase, UbiG +3.34 ± 1.26 
ABO_2036 Cytochrome c oxidase, subunit I, CyoB +5.75 ± 0.79 ABO_1238 Enoyl-CoA hydratase/isomerase, putative +4.79 ± 0.89 
ABO_1652 Fatty oxidation complex, FadB2 +11.01 ± 0.46 ABO_1653 3-oxoacyl-CoA thiolase, FadA2 +9.20 ± 0.60 
ABO_1701 Acyl-CoA dehydrogenase, C-terminal domain, AcdA +5.83 ± 0.87 ABO_2102 Acyl-CoA dehydrogenase, putative +9.03 ± 0.59 
ABO_1445 Sugar aldolase, putative +5.11 ± 0.86 ABO_2741 Isocitrate lyase +23.17 ± 0.20 
ABO_0211 Phosphoglucomutase/phosphomannomutase −4.37 ± 0.93 ABO_0426 Glycerophosphoryl diester phosphodiesterase precursor −5.95 ± 0.79 
ABO_1042 Glycerophosphoryl diester phosphodiesterase, putative −8.07 ± 0.52 ABO_2613 Phosphoglycerate kinase, pgk −4.38 ± 0.90 
ABO_2701 2′,3′-cyclic-nucleotide 2′-phosphodiesterase, CpdB −3.67 ± 1.15 ABO_0840 Adenylate kinase, Adk −8.47 ± 0.50 
ABO_0885 Adenylate cyclase −6.40 ± 0.69     
Stress-related genes 
ABO_2633 DNA-binding stress protein +3.95 ± 1.22 ABO_0314 Heat shock protein, DnaK +11.32 ± 0.39 
ABO_0315 Heat shock protein, DnaJ +14.67 ± 0.34 ABO_0633 Chaperonin, 10 kDa, GroES +9.17 ± 0.51 
ABO_0634 GroEL chaperonin, Cpn60 +17.84 ± 0.30 ABO_1180 Heat shock protein, HtpX +11.67 ± 0.54 
ABO_1489 Heat shock protein, HtpG +13.37 ± 0.42 ABO_1777 Heat-shock protein, IbpA +17.45 ± 0.27 
ABO_2244 Heat shock protein, HslV +11.67 ± 0.49 ABO_2245 Heat shock protein, HslU +9.72 ± 0.51 
ABO_0095 Lactoylglutathione lyase, GloA +10.22 ± 0.56 ABO_0649 Superoxide dismutase, SodB +9.99 ± 0.40 
ABO_1531 Protein-methionine-S-oxide reductase, MsrB +7.05 ± 0.70 ABO_2703 Peptide methionine sulfoxide reductase, MsrA +3.81 ± 1.13 
ABO_0758 TolB protein +6.86 ± 0.65     
Transport systems 
ABO_2119 Flavohemoprotein, HmpA +13.29 ± 0.41 ABO_2370 Fe3+-hydroxamate ABC transporter protein +6.60 ± 0.61 
ABO_2487 Ferripyoverdine receptor precursor, FpvB +6.97 ± 1.14 ABO_1360 Copper-translocating P-type ATPase +13.64 ± 0.34 
ABO_1364 Copper-binding protein, CopC +9.83 ± 0.57 ABO_1411 OmpA family protein +11.81 ± 0.46 
ABO_0618 Sodium/alanine symporter, Dag A −6.71 ± 0.61 ABO_0851 Nitrate transport ATP-binding protein, NtrC −3.40 ± 2.93 
ABO_0852 Nitrate ABC transporter permease protein, NtrB −3.93 ± 1.07 ABO_1668 Ion transport protein −4.61 ± 0.93 
ABO_1888 Sulfonate ABC transporter protein −5.45 ± 0.75 ABO_2509 Branched-chain amino acid ABC transporter −3.88 ± 1.23 
ABO_2512 Branched-chain amino acid ABC transporter, ATP-binding protein, putative −6.40 ± 0.89 ABO_2513 Amino acid ABC transporter, ATP-binding protein −5.35 ± 0.92 
ABO_2544 Small multidrug resistance protein −9.19 ± 0.48 ABO_2684 Phosphate ABC transporter permease protein −6.81 ± 0.85 
Cell wall metabolism 
ABO_0601 Cell division protein, FtsQ +3.84 ± 1.28 ABO_0603 Cell division protein, FtsZ +10.02 ± 0.45 
ABO_0120 Amidase, AmiC +6.19 ± 0.73 ABO_1957 Membrane-bound lytic murein transglycosylase B +4.60 ± 0.90 
ABO_0260 Effector of murein hydrolase, putative −4.00 ± 1.12 ABO_1221 Membrane-bound lytic murein transglycosylase D −7.82 ± 0.58 
ABO_1425 Soluble lytic transglycosylase, putative −11.06 ± 0.39 ABO_2215 Heptosyltransferase II, putative −11.43 ± 0.54 
Biofilm, formation of alginate 
ABO_0395 Mannose-6-phosphate isomerase, AlgA +12.31 ± 0.48 ABO_0126 Type-1 fimbrial protein, A chain precursor +15.45 ± 0.39 
ABO_0584 Phosphoheptose isomerase, GmhA +6.63 ± 0.62 ABO_0463 Pilus biogenesis protein, PilN −6.59 ± 0.62 
ABO_0467 Pilin biogenesis-related protein −10.02 ± 0.40 ABO_0613 Type IV pilus assembly protein, TapC −10.74 ± 0.53 
ABO_0614 IV fimbrial assembly protein, PilB −6.14 ± 0.74 ABO_2670 IV pili twitching motility protein, PilT −5.19 ± 0.78 
Transcriptional regulators 
ABO_0031 Transcriptional regulatory protein, Qseb +7.52 ± 0.60 ABO_0121 Transcriptional regulator, GntR family +9.19 ± 0.48 
ABO_1680 DnaK suppressor protein homolog, DksA1 +4.37 ± 0.93 ABO_1708 Sensor histidine kinase +11.58 ± 0.38 
ABO_1709 DNA-binding response regulator +10.82 ± 0.38 ABO_2132 HD-GYP domain protein +3.79 ± 1.26 
ABO_0005 DNA-binding protein H-NS −5.36 ± 0.76 ABO_0104 Two-component hybrid chemotactic sensor and regulator, ChpA, putative −7.52 ± 0.94 
ABO_0167 Phosphate regulon sensor protein, PhoR −7.48 ± 0.64 ABO_0428 DNA-binding response regulator, LuxR family −11.76 ± 0.35 
ABO_0472 Two-component sensor histidine kinase, PilS −8.24 ± 0.49 ABO_0859 Nitrogen assimilation response regulatory protein −7.40 ± 0.63 
ABO_1745 Stringency control −6.78 ± 0.65 ABO_1840 Stringency control −11.76 ± 0.34 
ABO_1986 Two-component sensor histidine kinase protein, putative −7.13 ± 1.02 ABO_2018 Sensor histidine kinase/response regulator −10.05 ± 0.53 
ABO_2023 Catabolite gene activator, Crp −9.52 ± 0.52 ABO_2260 Sensory box histidine kinase, NtrB −6.55 ± 0.61 
ABO_2433 Conserved hypothetical signal transduction protein −7.14 ± 0.55 ABO_2508 Transcriptional regulator, GntR family −16.05 ± 0.35 
ABO_2519 Response regulator −11.50 ± 0.37 ABO_2520 Sensor histidine kinase −13.98 ± 0.29 
ABO_2530 Nitrogen regulatory protein P-II, GlnK2 −7.49 ± 0.56 ABO_2581 Transcription regulator family protein, AcrR −8.39 ± 0.51 
*

Gene numbers, gene function, and functional category are presented according to the annotated genome (Schneiker, 2006).

ER, expression ratio; SD, standard deviation.

1

 Combined schematic representation of genomic regions containing genes encoding monooxygenases and enzymes presumably involved in the initial oxidation of alkanes derived from the analysis of proteomic (Sabirova, 2006) and transcriptomic data. Genes upregulated in alkane-grown cells are shown in black; white-colored ORFs with black frames show genes in the Alcanivorax borkumensis genome that seem not to be upregulated; grey-colored ORFs show the P450-1 and P450-2 genes (ABO_0201 and ABO_2288, correspondingly), encoding identical p450 cytochrome proteins.

1

 Combined schematic representation of genomic regions containing genes encoding monooxygenases and enzymes presumably involved in the initial oxidation of alkanes derived from the analysis of proteomic (Sabirova, 2006) and transcriptomic data. Genes upregulated in alkane-grown cells are shown in black; white-colored ORFs with black frames show genes in the Alcanivorax borkumensis genome that seem not to be upregulated; grey-colored ORFs show the P450-1 and P450-2 genes (ABO_0201 and ABO_2288, correspondingly), encoding identical p450 cytochrome proteins.

2

 Representation of the metabolic response of Alcanivorax borkumensis SK2 to alkane exposure as opposed to pyruvate. Genes repressed or induced after exposure to alkanes are shown with a grey or a red background, respectively.

2

 Representation of the metabolic response of Alcanivorax borkumensis SK2 to alkane exposure as opposed to pyruvate. Genes repressed or induced after exposure to alkanes are shown with a grey or a red background, respectively.

Alkane-induced biofilm formation and adhesion to hydrocarbons

Attachment of A. borkumensis to hydrocarbons and its molecular mechanisms have not yet been studied, although such abilities are likely to form part of the specific ecological adaptation of this bacterium. EM observation of Alcanivorax SK2 indeed indicates that this organism forms biofilm-supporting structures during growth on alkanes (Fig. 3). Cells grown on alkane seem to more connect to each other rather than to the solid surface of the carrier slide, and they are shorter and rounder, and produce considerable amount of extracellular polymeric substances (EPS), which appears to support the three-dimensional structure of a biofilm. After 10 days of growth, alkane-grown cells develop a biofilm, which exhibits a pronounced three-dimensional architecture supported by extracellular matrix (Fig. 3).

3

 Transmission electron microscopy observations of Alcanivorax borkumensis SK2 cells, their adhesion, and biofilm formation on abiotic surfaces. The cells were grown on either pyruvate (a, b) or hexadecane (c, d) on Permanox slides in ONR7a medium. The biofilms were monitored after 4 days (a, c) and 10 days (b, d) of growth.

3

 Transmission electron microscopy observations of Alcanivorax borkumensis SK2 cells, their adhesion, and biofilm formation on abiotic surfaces. The cells were grown on either pyruvate (a, b) or hexadecane (c, d) on Permanox slides in ONR7a medium. The biofilms were monitored after 4 days (a, c) and 10 days (b, d) of growth.

The argument of an alkane-induced formation of EPS is supported by alkane-induced up-expression of gmhA (ABO_0584). GmhA encodes a phosphoheptose isomerases that mediates the synthesis of heptose, a conserved component of outer membrane lipopolysaccharide, that for example in Yersinia, was shown to contribute to the formation of biofilms (Darby, 2005). Another indication for alkane-induced biosynthesis of EPS in Alcanivorax could be the upregulation of mannose-6-phosphate isomerase algA, encoded by ABO_0395 (Table 1), predicted to encode an enzyme known to catalyze the first and third steps of the biosynthetic pathway of alginate, a known component of biofilms in other organisms (Shinabarger, 1991); however, algA, like other related alg genes, may also be involved in the biosynthesis of lipopolysaccharide, as shown elsewhere (Goldberg, 1993; Gaona, 2004). As direct experimental evidence is still missing as regards alginate production by Alcanivorax, our findings may point more to the biosynthesis of lipopolysaccharide, rather than of alginate. While the biosynthesis of alginate has not yet been shown for Alcanivorax, it has been described for marine algae and bacteria belonging to the genera Pseudomonas and Azotobacter (Gorin & Spencer, 1966; Lin & Hassid, 1966; Evans & Linker, 1973).

With respect to the expression of genes thought to be involved in the signalling and regulatory processes essential for the formation of biofilm, our transcriptomic data show that the widely suggested mechanism of biofilm formation mediated by elevated concentrations of messenger c-di-GMP does not seem to be clearly effective in the case of Alcanivorax growing on alkanes. One regulatory system (encoded by ABO_2433), containing the known GGDEF and EAL domains, responsible for the biosynthesis and hydrolysis of c-di-GMP, respectively, was found to be downregulated, while another gene, ABO_2132 encoding the HD-GYP domain with phosphodiesterase activity responsible for hydrolysis of c-di-GMP (Galperin, 1999; Ryan, 2006), was found to be upregulated. Hence, the precise role of intracellular c-di-GMP levels for biofilm formation may be more complex than previously assumed (Hickman, 2005; Römling, 2005). We furthermore found that a whole set of genes involved in the formation of pili (ABO_0463, ABO_0467, ABO_0613, ABO_00614, and ABO_2670, Table 1) is downregulated during growth on alkanes; hence, attachment of Alcanivorax to alkane droplets does seem to require quorum sensing, and leads to enhanced biosynthesis of EPS, and yet, it may not be classical biofilms that are formed to access alkane droplets, triggered by intracellular c-di-GMP and by the formation of pili and/or fimbriae, but rather irregular aggregates glued together by extracellular polysaccharides.

Transport systems

Our expression data also shed some new light on the acknowledged uncertainty as to how alkanes are transported into the bacterial cell. One of the alkane-induced genes, ABO_0902 (see Supporting Information, Table S2), encoding an outer membrane lipoprotein Blc, may directly be involved in the alkane uptake process, as it contains a so-called lipocalin domain, which, in contact with organic solvents, forms a small hydrophobic pocket catalyzing the transport of small hydrophobic molecules, such as lipids and steroid hormones (Pervaiz & Brew, 1987). This alkane-induced protein would thus be a prime candidate potentially mediating alkane transport.

Regulation

Using a transcriptomics approach, a number of additional alkane-induced regulatory systems have been detected (Table 1), as compared with our previous proteomics study (Sabirova, 2006). A transcriptional regulator of the GntR family, encoded by ABO_0121, is located next to the ABO_0122 encoding the alkB2 monooxygenase, suggesting that the ABO_0121-encoded gene product might regulate the expression of the adjacent monooxygenase. Another regulatory system consisting of ABO_1708 and ABO_1709, adjacent to each other and likely to be operon-arranged, encodes a pair of sensor histidine kinase and DNA-binding response regulator that are also upregulated on alkanes. Their close proximity to the gene of fatty acid degradation (fadH dienoyl-CoA reductase) may indicate that this regulatory system controls the oxidation of fatty acids in Alcanivorax. Our transcriptome data also hint towards quorum sensing playing a role in biofilm formation of Alcanivorax on alkanes, as the major transcriptional regulator QseB encoded by ABO_0031 was found to be upregulated on hexadecane (Table 1). Quorum sensing has indeed been reported to trigger biofilm formation via the biosynthesis of extracellular exopolysaccharides (EPS) (Sauer, 2002), also visible on our EM pictures. We did not detect increased expression of the cognate histidine kinase, QseC, encoded by ABO_0030. This finding indicates that for initial signal reception and transduction constant levels of sensor protein suffice, while the subsequent coordinated regulation of the expanded quorum-sensing regulon qse does require increased titers of Qse regulator protein. Finally, an HD-GYP domain protein encoded by ABO_2132 and mentioned earlier in ‘Alkane-induced biofilm formation and adhesion to hydrocarbons’ is also upregulated on alkanes and hence represents another worthy target for regulatory studies of growth on alkanes.

To conclude, our transcriptomics analysis of A. borkumensis responses to alkane exposure adds a complementary view on alkane metabolism by this bacterium, in addition to our previous proteomics study, and reveals a number of novel observations, for instance concerning the molecular mechanisms of alkane transport across the cytoplasmic membrane, and pointing to a diverse set of enzymes for the degradation of alkanes. Alcanivorax SK2 seems to respond to growth on alkanes by forming cell aggregates, probably supported by enhanced synthesis of EPS and probably following in a quorum-sensing-mediated aggregation process. Finally, the study has also revealed many transcriptional regulators to be differentially expressed, indicating a complex regulatory interplay of alkane degradation with other metabolic functions in this marine organism. Knocking out these alkane-responsive regulators should provide a valuable insight into the scope of regulons and subregulons essential for alkane degradation by A. borkumensis SK2.

Acknowledgements

This research was supported by a grant from the German Ministry for Education and Research (BMBF) in the frame of the GenoMik network ‘Genome Research on Bacteria Relevant for Agriculture, Environment and Biotechnology’ and by a short-term fellowship from the European Molecular Biology Organization (EMBO) (ASTF 354-2006).

References

Brune
I
Becker
A
Paarmann
D
Albersmeier
A
Kalinowski
J
Pühler
A
Tauch
A
(
2006
)
Under the influence of the active deodorant ingredient 4-hydroxy-3-methoxybenzyl alcohol, the skin bacterium Corynebacterium jeikeium moderately responds with differential gene expression
.
J Biotechnol
 
127
:
21
33
.
Darby
C
Ananth
SL
Tan
L
Hinnebusch
BJ
(
2005
)
Identification of gmhA, a Yersinia pestis gene required for flea blockage, by using a Caenorhabditis elegans biofilm system
.
Infect Immun
 
73
:
7236
7242
.
Dondrup
M
Goesmann
A
Bartels
D
Kalinowski
J
Krause
L
Linke
B
Rupp
O
Sczyrba
A
Pühler
A
Meyer
F
(
2003
)
EMMA: a platform for consistent storage and efficient analysis of microarray data
 . J Biotechnol
106
:
135
146
.
Dudoit
S
Yang
YH
Callow
MJ
Speed
TP
(
2002
)
Statistical methods for identifying differentially expressed genes in replicated cDNA microarray experiments
.
Stat Sinica
 
12
:
111
139
.
Dutta
TK
Harayama
S
(
2001
)
Biodegradation of n-alkylcycloalkanes and n-alkylbenzenes via new pathways in Alcanivorax sp. strain MBIC 4326
.
Appl Environ Microb
 
67
:
1970
1974
.
Evans
LR
Linker
A
(
1973
)
Production and characterization of the slime polysaccharide of Pseudomonas aeruginosa
.
J Bacteriol
 
116
:
915
924
.
Galperin
MY
Natale
DA
Aravind
L
Koonin
EV
(
1999
)
A specialized version of the HD hydrolase domain implicated in signal transduction
.
J Mol Microb Biotech
 
1
:
303
305
.
Gaona
G
Núñez
C
Goldberg
JB
Linford
AS
Nájera
R
Castañeda
M
Guzmán
J
Espín
G
Soberón-Chávez
G
(
2004
)
Characterization of the Azotobacter vinelandii algC gene involved in alginate and lipopolysaccharide production
.
FEMS Microbiol Lett
 
238
:
199
206
.
Gertler
C
Gerdts
G
Timmis
KN
Yakimov
MM
Golyshin
PN
(
2009
)
Populations of heavy fuel oil-degrading marine microbial community in presence of oil sorbent materials
.
J Appl Microbiol
 
107
:
590
605
.
Goldberg
JB
Hatano
K
Pier
GB
(
1993
)
Synthesis of lipopolysaccharide-O-side-chains by Pseudomonas aeruginosa PAO1 requires the enzyme phosphomannomutase
.
J Bacteriol
 
175
:
1605
1611
.
Gorin
PA
Spencer
JF
(
1966
)
Exocellular alginic acid from Azotobacter vinelandii
.
Can J Chem
 
44
:
993
998
.
Hickman
JW
Tifrea
DF
Harwood
CS
(
2005
)
A chemosensory system that regulates biofilm through modulation of cyclic diguanylate levels
.
P Natl Acad Sci USA
 
102
:
14422
14427
.
Lin
TY
Hassid
WZ
(
1966
)
Pathway of alginic acid synthesis in the marine brown alga, Fucus gardneri Silva
.
J Biol Chem
 
24
:
5284
5297
.
Lünsdorf
H
Strompl
C
Osborn
AM
Bennasar
A
Moore
ER
Abraham
WR
Timmis
KN
(
2001
)
Approach to analyse interactions of microorganisms, hydrophobic substrates and soil colloids leading to formation of composite biofilms, and to study initial events in microbiogeological processes
.
Method Enzymol
 
336
:
317
331
.
McKew
BA
Coulon
F
Yakimov
MM
Denaro
R
Genovese
M
Smith
CJ
Osborn
AM
Timmis
KN
McGenity
TJ
(
2007
)
Efficacy of intervention strategies for bioremediation of crude oil in marine systems and effects on indigenous hydrocarbonoclastic bacteria
.
Environ Microbiol
 
9
:
1562
1571
.
Palackal
NT
Burczynski
ME
Harvey
RG
Pennig
TM
(
2001
)
Metabolic activation of polycyclic aromatic hydrocarbon trans-dihydrodiols by ubiquitously expressed aldehyde reductase (AKR1A1)
.
Chem Biol Interact
 
130–132
:
815
824
.
Pervaiz
S
Brew
K
(
1987
)
Homology and structure-function correlations between alpha 1-acid glycoprotein and serum retinol-binding protein and its relatives
.
FASEB J
 
1
:
209
214
.
Römling
U
Gomelsky
M
Galperin
MY
(
2005
)
C-di-GMP: the dawning of a novel bacterial signalling system
 . Mol Microbiol
57
:
629
639
.
Ryan
RP
Fouhy
Y
Lucey
JF
et al.  . (
2006
)
Cell-cell signaling in Xanthomonas campestris involves an HD-GYP domain protein that functions in cyclic di-GMP turnover
.
P Natl Acad Sci USA
 
103
:
6712
6717
.
Sabirova
JS
Ferrer
M
Regenhardt
D
Timmis
KN
Golyshin
PN
(
2006
)
Proteomic insights into metabolic adaptations in Alcanivorax borkumensis induced by alkane utilization
.
J Bacteriol
 
188
:
3763
3773
.
Sauer
K
Camper
A
Ehrlich
G
Costerton
J
Davies
D
(
2002
)
Pseudomonas aeruginosa displays multiple phenotypes during development as a biofilm
.
J Bacteriol
 
184
:
1140
1154
.
Schneiker
S
Martins dos Santos
VA
Bartels
D
et al.  . (
2006
)
Genome sequence of the ubiquitous hydrocarbon-degrading marine bacterium Alcanivorax borkumensis
.
Nat Biotechnol
 
24
:
997
1004
.
Shinabarger
D
Berry
A
May
TB
Rothmel
R
Fialho
A
Chakrabarty
AM
(
1991
)
Purification and characterization of phosphomannose isomerase-guanosine diphospho-d-mannose pyrophosphorylase – a bifunctional enzyme in the alginate biosynthetic pathway of Pseudomonas aeruginosa
.
J Biol Chem
 
29
:
2080
2088
.
Van Beilen
JB
Funhoff
EG
(
2007
)
Alkane hydroxylases involved in microbial alkane degradation
.
Appl Microbiol Biot
 
74
:
13
21
.
Van Beilen
JB
Wubbolts
MG
Witholt
B
(
1994
)
Genetics of alkane oxidation by Pseudomonas oleovorans
.
Biodegradation
 
5
:
161
174
.
Wentzel
A
Ellingsen
TE
Kotlar
H-K
Zotchev
SB
Throne-Holst
M
(
2007
)
Bacterial metabolism of long-chain n-alkanes
.
Appl Microbiol Biot
 
76
:
1209
1221
.
Yakimov
MM
Timmis
KN
Golyshin
PN
(
2007
)
Obligate oil-degrading marine bacteria
.
Curr Opin Biotech
 
18
:
257
266
.
Yang
YH
Dudoit
S
Luu
P
Lin
DM
Peng
V
Ngai
J
Speed
TP
(
2002
)
Normalization for cDNA microarray data: a robust composite method addressing single and multiple slide systematic variation
 . Nucleic Acids Res
30
:
e15
.

Supporting Information

Please note: Wiley-Blackwell is not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

Author notes

Editor: Michael Galperin