Gene expression of terminal oxidases in two marine bacterial strains exposed to nanomolar oxygen concentrations

Gene expression of terminal oxidases in two marine bacterial strains exposed to nanomolar oxygen... Abstract The final step of aerobic respiration is carried out by a terminal oxidase transporting electrons to oxygen (O2). Prokaryotes harbor diverse terminal oxidases that differ in phylogenetic origin, structure, biochemical function, and affinity for O2. Here we report on the expression of high-affinity (cytochrome cbb3 oxidase), low-affinity (cytochrome aa3 oxidase), and putative low-affinity (cyanide-insensitive oxidase (CIO)) terminal oxidases in the marine bacteria Idiomarina loihiensis L2-TR and Marinobacter daepoensis SW-156 upon transition to very low O2 concentrations (<200 nM), measured by RT-qPCR. In both strains, high-affinity cytochrome cbb3 oxidase showed the highest expression levels and was significantly up-regulated upon transition to low O2 concentrations. Low-affinity cytochrome aa3 oxidase showed very low transcription levels throughout the incubation. Surprisingly, however, it was also up-regulated upon transition to low O2 concentrations. In contrast, putative low-affinity CIO had much lower expression levels and markedly different regulation patterns between the two strains. These results demonstrate that exposure to low O2 concentrations regulates the gene expression of different types of terminal oxidases, but also that the type and magnitude of transcriptional response is species-dependent. Therefore, in situ transcriptome data cannot, without detailed knowledge of the transcriptional regulation of the species involved, be translated into relative respiratory activity. nanomolar oxygen, gene expression, terminal oxidase, RT-qPCR, marine bacteria INTRODUCTION Oxygen (O2) is the most favourable electron acceptor in microbial respiration (Canfield, Kristensen and Thamdrup 2005). It may therefore not seem surprising that O2 can be used as a terminal electron acceptor even at nanomolar concentrations (Stolper, Revsbech and Canfield 2010). This is ecologically relevant in many micro-oxic environments, e.g. in marine oxygen minimum zones (OMZs). OMZs are oceanic areas with a layer of seawater at intermediate depth with an O2 concentration <20 µmol l−1 (Helly and Levin 2004), sometimes fluctuating from a few micromolar to as low as a few nanomolar. OMZs have a global impact on the biogeochemistry of the oceans, especially with respect to nitrogen cycling (Wright, Konwar and Hallam 2012), and show high rates of aerobic respiration despite low O2 concentrations (Garcia-Robledo et al. 2017). The biochemical basis for this are terminal oxidases with very high O2 affinities, i.e. with half-saturation constants (Km) in the range of 3–8 nmol l−1 (D'mello, Hill and Poole 1996). Terminal oxidases transfer electrons to O2 in the final step of aerobic respiration (Poole 1983), and have been classified into two main families according to their structural and functional differences (Calhoun, Thomas and Gennis 1994; Pitcher and Watmough 2004; Brändén, Gennis and Brzezinski 2006; Hemp et al. 2006): (i) the heme-copper oxidases (HCOs), which can be subdivided into A-class HCOs (with low O2 affinity) and B- and C-class HCOs (with high O2 affinity) (Morris and Schmidt 2013); and ii) the cytochrome bd quinol oxidases, which can be subdivided into classical bd-type (with high O2 affinity) (Borisov et al. 2011) and cyanide-insensitive oxidases (CIOs: putative with low O2 affinity) (Jackson et al. 2007). CIOs are homologs of classical cytochrome bd oxidases, but are phylogenetically distinct (Miura et al. 2013). Kinetic analysis showed that CIOs in diverse bacteria, e.g. Pseudomonas aeruginosa (Arai et al. 2014), Campylobacter jejuni (Jackson et al. 2007) and Gluconobacter oxydans (Miura et al. 2013), have low affinity for O2, even comparable to low-affinity A-class HCOs. O2 concentration has been shown to regulate terminal oxidase expression in a few bacterial species containing both high- and low-affinity enzymes (e.g. P. aeruginosa (Alvarez-Ortega and Harwood 2007; Arai 2011; Arai et al. 2014), Desulfovibrio vulgaris (Lamrabet et al. 2011), and Escherichia coli (Cotter et al. 1990)), albeit never under nanomolar O2 concentrations. Environmental data suggest that genes for both high- and low-affinity terminal oxidases are widely distributed and simultaneously expressed in diverse environments (Morris and Schmidt 2013), including OMZs under varying O2 concentrations (Kalvelage et al. 2015). The link between O2 concentration, terminal oxidase expression, and the kinetics of O2 respiration is thus not well understood, in particular not at nanomolar O2 concentrations. Only recently, novel sensor developments (Revsbech et al. 2009; Lehner et al. 2015) have enabled detailed measurements of O2 respiration rates and kinetics at nanomolar O2 concentrations, both in situ (Garcia-Robledo et al. 2017) and in controlled pure culture incubations (Gong et al. 2016). The latter study followed the change in O2 respiration kinetics of four marine bacterial strains during the transition to nanomolar O2 concentrations and revealed apparent Km values of 10–60 nmol l−1 (Gong et al. 2016). The aim of the present study was to follow up on this kinetic study (Gong et al. 2016) by quantifying the expression of both low- and high-affinity terminal oxidases during the transition from aerobic to nearly anoxic conditions, in order to link gene expression with respiration kinetics. To achieve this we selected two of the investigated strains (Idiomarina loihiensis L2-TR DSM-15 497 and Marinobacter daepoensis SW-156 DSM-16 072) that (i) carry single-copy genes of high-affinity terminal oxidase (cytochrome cbb3 oxidase, C-class HCO), low-affinity terminal oxidase (cytochrome aa3 oxidase, A-class HCO) and putative low-affinity terminal oxidase (cytochrome bd quinol oxidase, CIO; Fig. S1); and (ii) were previously detected in aquatic environments (Fig. S2), including OMZs (Table S1). Terminal oxidase genes and transcripts were detected by reverse transcription quantitative PCR (RT-qPCR) assays specifically developed for all individual genes and strains. MATERIALS AND METHODS Bacterial cultures I. loihiensis L2-TR DSM-15 497 and M. daepoensis SW-156 DSM-16 072 were obtained from the Leibniz Institute DSMZ, Germany. Bacteria were grown at 25°C in autoclaved marine broth media containing 5 g peptone (BD), 1 g yeast extract (Difco), 0.1 g FeC6H5O7, 12.6 g MgCl2·6 H2O, 3.24 g Na2SO4, 19.45 g NaCl, 2.38 g CaCl2·2H2O, 0.55 g KCl, 0.16 g NaHCO3 and 0.01 g Na2HPO4·2H2O l-1. Bacterial cultures were inoculated into fresh marine broth media and incubated overnight on a rotary shaker (180 rpm) to ensure fully oxic conditions and also obtain cell suspensions in the exponential growth phase. Set-up and incubation The highly sensitive optical O2 sensors used in this study have been described previously (Borisov, Lehner and Klimant 2011; Lehner et al. 2015; Gong et al. 2016); in brief, combinations of a fluorophore and immobilizing matrix resulted in two types of O2 optodes: a highly sensitive optode with a measuring range of 1–1000 nmol l−1, and a broader range optode with a measuring range of 0.01–10 µmol l−1. Data were recorded every 10 s during the incubations. Incubation bottles (2 l; Schott) with pre-glued optode dots of both types inside, and a glass-coated magnet in the bottom, were filled with 1.5 l of marine broth medium. Optode dots were positioned 2–3 cm above the bottom to ensure that they were immersed in the liquid phase during the incubation. Two glass tubes went through the butyl rubber bottle cap for flushing the medium with N2 to remove dissolved O2 and to take samples. All bottles, glass-coated magnets and tubes were previously cleaned with 0.1 mol l−1 HCl, and subsequently washed with autoclaved water to minimize bacterial contamination. The optodes and permanently attached optoelectronics (Lehner et al. 2015; Gong et al. 2016) were calibrated before the incubation by addition of various volumes of air-saturated water to freshly deoxygenated water. Incubations were performed in triplicate in a constant-temperature water bath at 25°C during the experiments. Before injecting bacteria, the bottles with media were kept in the water bath for 30 min to reach temperature equilibrium. The incubations lasted for 300–350 min. The O2 concentration was controlled by injecting air-equilibrated artificial seawater or flushing with N2 during the incubation. In order to estimate the growth phase of bacteria in the inoculum, the optical density at 600 nm was measured with a spectrophotometer (Pharmacia Biotech). The value of OD600 of the stock culture was adjusted to 0.5 with marine broth medium before inoculation (Table S2). A volume of 5 ml of the adjusted bacterial inoculum was injected into the medium contained in the bottle when the O2 concentration decreased to 10 µmol l−1 by N2 bubbling. The first sample was taken once the bacteria were injected into the incubation bottle. The bubbling with N2 was terminated after the inoculation, so any further decreases in O2 concentration were due to respiration. The second sample was taken when the O2 concentration had decreased to 5 µmol l−1. Time zero in the subsequent annotation was defined as the time point when the O2 concentration had decreased to 200 nmol l−1. Further samples were taken when the cultures had been exposed to low O2 concentrations (≤ 200 nmol l−1) for 5, 10, 30, 120 and 200 min. Samples of 100 ml were taken by a syringe connected to one of the two glass tubes penetrating the stopper. Samples were filtered onto 0.22-µm pore-size polycarbonate filters (47 mm, Sigma), and immediately transferred to micro-centrifuge tubes containing 250 µl water and 500 µl RNAprotect Bacteria Reagent (Qiagen). All processes from sampling to immersion in RNAprotect Bacteria Reagent were finished within 5 min. The mixture of filter and reagent was vortexed for 5 s and subsequently incubated for 30 min at room temperature. After centrifuging for 10 min at 5000 × g, the supernatant was removed. The samples were stored at −80°C. Nucleic acid extraction and preparation of cDNA Cells on the filter were lysed by the lysozyme and proteinase K method described in the RNeasy Mini kit (Qiagen). Half of the lysate was used for DNA extraction by the DNeasy Blood and Tissue kit (Qiagen) following the manufacturer's protocol. The quality of the DNA extracts was checked by using serial dilutions as templates for qPCR to confirm that the extracts did not contain inhibitors for qPCR. The other half of the cell lysate was used for RNA extraction by the RNeasy Mini kit (Qiagen) according to the manufacturer's instructions. Genomic DNA co-extracted during RNA extraction was removed by TURBO DNA-free kit (Ambion), and the complete removal was confirmed by qPCR of a terminal oxidase gene and a reference gene (see below), both of which were negative for RNA extracts. RNA was checked for integrity on 0.8% agarose electrophoresis gels, quantified by Qubit RNA BR Assay Kit (Thermo Fisher Scientific), and reverse-transcribed using the iScript cDNA Synthesis kit (Bio-rad) following the manufacturer's instructions. The cDNA was stored at −20°C. Primer design, PCR, and qPCR standard The genes of subunit I of each terminal oxidase and of rpoB, encoding the RNA polymerase β subunit (as reference gene), were retrieved from NCBI for I. loihiensis and M. daepoensis and aligned in Geneious 4.8.5. Specific PCR primers (Tables 1 and S3) were designed manually and checked in silico by OligoAnalyzer 3.1 (http://eu.idtdna.com/calc/analyzer). Primers and PCR conditions for the 16S rRNA gene (Tables S3 and S4) were as previously described (Lever et al. 2015). Table 1. Target genes and specific qPCR assays for detecting terminal oxidases in I. loihiensis and M. daepoensis. Species Name I. loihiensis L2-TR DSM-15 497 M. daepoensis SW-156 DSM-16 072 Target protein 1(1) cbb3-type cytochrome c oxidase subunit 1 cbb3-type cytochrome c oxidase subunit 1 Oxidase type (gene) C-class HCO (fixN) C-class HCO (fixN) Oxidase affinity High affinity High affinity Locus tag (source) IL1297 (NCBI) K325DRAFT_1727 (IMG) Primer sequence; name CTGACCGCTGGCTTCCTGGG; Il-fixNF CTGACCGCCGGCTTCCTGGG; Md-fixNF (5′→3′) GTCATGATACCGTTAATCATGCCGCCCCA; Il-fixNR GTCATCATACCGTTGATCATGCCGCCCCA; Md-fixNR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 82 86 Target protein 2 cytochrome aa3 oxidase subunit I cytochrome aa3 oxidase subunit I Oxidase type (gene) A-class HCO (ctaD) A-class HCO (ctaD) Oxidase affinity Low affinity Low affinity Locus tag (source) IL0259 (NCBI) K325DRAFT_2803 (IMG) Primer sequence; name TTCGGTACGAGTTTCTTTGACGCC; Il-ctaDF TTCGGGACAAGTTTCTTTGACGCC; Md-ctaDF (5′→3′) AGTTAAAGATCTTCACCCCGGTGGG; Il-ctaDR AGTTAAACACCTTCACCCCGGTGGG; Md-ctaDR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 83 86 Target protein 3 cytochrome bd-type quinol oxidase subunit 1 cytochrome bd-I ubiquinol oxidase subunit 1 apoprotein Oxidase type (gene) CIO (cydA) CIO (cydA) Oxidase affinity Putative low-affinity Putative low-affinity Locus tag (source) IL0041 (NCBI) K325DRAFT_1300 (IMG) Primer sequence; name GCGGTGGGCACGGCTATTTCAGC; Il-cydAF GCCACGGGTACCTTTATTTCCTC; Md-cydAF (5′→3′) TTTTACGGGCGGCGGCGTTGGT; Il-cydAR CCTTACGGTTTGCCTCTACCTCA; Md-cydAR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 86 86 Species Name I. loihiensis L2-TR DSM-15 497 M. daepoensis SW-156 DSM-16 072 Target protein 1(1) cbb3-type cytochrome c oxidase subunit 1 cbb3-type cytochrome c oxidase subunit 1 Oxidase type (gene) C-class HCO (fixN) C-class HCO (fixN) Oxidase affinity High affinity High affinity Locus tag (source) IL1297 (NCBI) K325DRAFT_1727 (IMG) Primer sequence; name CTGACCGCTGGCTTCCTGGG; Il-fixNF CTGACCGCCGGCTTCCTGGG; Md-fixNF (5′→3′) GTCATGATACCGTTAATCATGCCGCCCCA; Il-fixNR GTCATCATACCGTTGATCATGCCGCCCCA; Md-fixNR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 82 86 Target protein 2 cytochrome aa3 oxidase subunit I cytochrome aa3 oxidase subunit I Oxidase type (gene) A-class HCO (ctaD) A-class HCO (ctaD) Oxidase affinity Low affinity Low affinity Locus tag (source) IL0259 (NCBI) K325DRAFT_2803 (IMG) Primer sequence; name TTCGGTACGAGTTTCTTTGACGCC; Il-ctaDF TTCGGGACAAGTTTCTTTGACGCC; Md-ctaDF (5′→3′) AGTTAAAGATCTTCACCCCGGTGGG; Il-ctaDR AGTTAAACACCTTCACCCCGGTGGG; Md-ctaDR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 83 86 Target protein 3 cytochrome bd-type quinol oxidase subunit 1 cytochrome bd-I ubiquinol oxidase subunit 1 apoprotein Oxidase type (gene) CIO (cydA) CIO (cydA) Oxidase affinity Putative low-affinity Putative low-affinity Locus tag (source) IL0041 (NCBI) K325DRAFT_1300 (IMG) Primer sequence; name GCGGTGGGCACGGCTATTTCAGC; Il-cydAF GCCACGGGTACCTTTATTTCCTC; Md-cydAF (5′→3′) TTTTACGGGCGGCGGCGTTGGT; Il-cydAR CCTTACGGTTTGCCTCTACCTCA; Md-cydAR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 86 86 (1): For ease of use, the same gene symbols coding for the terminal oxidases are used for these two species throughout the study, i.e. target 1: fixN; target 2: ctaD; and target 3: cydA. The rule for this modification was that those genes with different gene symbols coding for the same functional subunit are based on the BLAST annotation and sequence. The detailed primer sequences, annealing temperatures and detection temperatures in qPCR for rpoB and 16S rRNA genes are listed in the supplementary material (Tables S2 and S3). View Large Table 1. Target genes and specific qPCR assays for detecting terminal oxidases in I. loihiensis and M. daepoensis. Species Name I. loihiensis L2-TR DSM-15 497 M. daepoensis SW-156 DSM-16 072 Target protein 1(1) cbb3-type cytochrome c oxidase subunit 1 cbb3-type cytochrome c oxidase subunit 1 Oxidase type (gene) C-class HCO (fixN) C-class HCO (fixN) Oxidase affinity High affinity High affinity Locus tag (source) IL1297 (NCBI) K325DRAFT_1727 (IMG) Primer sequence; name CTGACCGCTGGCTTCCTGGG; Il-fixNF CTGACCGCCGGCTTCCTGGG; Md-fixNF (5′→3′) GTCATGATACCGTTAATCATGCCGCCCCA; Il-fixNR GTCATCATACCGTTGATCATGCCGCCCCA; Md-fixNR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 82 86 Target protein 2 cytochrome aa3 oxidase subunit I cytochrome aa3 oxidase subunit I Oxidase type (gene) A-class HCO (ctaD) A-class HCO (ctaD) Oxidase affinity Low affinity Low affinity Locus tag (source) IL0259 (NCBI) K325DRAFT_2803 (IMG) Primer sequence; name TTCGGTACGAGTTTCTTTGACGCC; Il-ctaDF TTCGGGACAAGTTTCTTTGACGCC; Md-ctaDF (5′→3′) AGTTAAAGATCTTCACCCCGGTGGG; Il-ctaDR AGTTAAACACCTTCACCCCGGTGGG; Md-ctaDR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 83 86 Target protein 3 cytochrome bd-type quinol oxidase subunit 1 cytochrome bd-I ubiquinol oxidase subunit 1 apoprotein Oxidase type (gene) CIO (cydA) CIO (cydA) Oxidase affinity Putative low-affinity Putative low-affinity Locus tag (source) IL0041 (NCBI) K325DRAFT_1300 (IMG) Primer sequence; name GCGGTGGGCACGGCTATTTCAGC; Il-cydAF GCCACGGGTACCTTTATTTCCTC; Md-cydAF (5′→3′) TTTTACGGGCGGCGGCGTTGGT; Il-cydAR CCTTACGGTTTGCCTCTACCTCA; Md-cydAR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 86 86 Species Name I. loihiensis L2-TR DSM-15 497 M. daepoensis SW-156 DSM-16 072 Target protein 1(1) cbb3-type cytochrome c oxidase subunit 1 cbb3-type cytochrome c oxidase subunit 1 Oxidase type (gene) C-class HCO (fixN) C-class HCO (fixN) Oxidase affinity High affinity High affinity Locus tag (source) IL1297 (NCBI) K325DRAFT_1727 (IMG) Primer sequence; name CTGACCGCTGGCTTCCTGGG; Il-fixNF CTGACCGCCGGCTTCCTGGG; Md-fixNF (5′→3′) GTCATGATACCGTTAATCATGCCGCCCCA; Il-fixNR GTCATCATACCGTTGATCATGCCGCCCCA; Md-fixNR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 82 86 Target protein 2 cytochrome aa3 oxidase subunit I cytochrome aa3 oxidase subunit I Oxidase type (gene) A-class HCO (ctaD) A-class HCO (ctaD) Oxidase affinity Low affinity Low affinity Locus tag (source) IL0259 (NCBI) K325DRAFT_2803 (IMG) Primer sequence; name TTCGGTACGAGTTTCTTTGACGCC; Il-ctaDF TTCGGGACAAGTTTCTTTGACGCC; Md-ctaDF (5′→3′) AGTTAAAGATCTTCACCCCGGTGGG; Il-ctaDR AGTTAAACACCTTCACCCCGGTGGG; Md-ctaDR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 83 86 Target protein 3 cytochrome bd-type quinol oxidase subunit 1 cytochrome bd-I ubiquinol oxidase subunit 1 apoprotein Oxidase type (gene) CIO (cydA) CIO (cydA) Oxidase affinity Putative low-affinity Putative low-affinity Locus tag (source) IL0041 (NCBI) K325DRAFT_1300 (IMG) Primer sequence; name GCGGTGGGCACGGCTATTTCAGC; Il-cydAF GCCACGGGTACCTTTATTTCCTC; Md-cydAF (5′→3′) TTTTACGGGCGGCGGCGTTGGT; Il-cydAR CCTTACGGTTTGCCTCTACCTCA; Md-cydAR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 86 86 (1): For ease of use, the same gene symbols coding for the terminal oxidases are used for these two species throughout the study, i.e. target 1: fixN; target 2: ctaD; and target 3: cydA. The rule for this modification was that those genes with different gene symbols coding for the same functional subunit are based on the BLAST annotation and sequence. The detailed primer sequences, annealing temperatures and detection temperatures in qPCR for rpoB and 16S rRNA genes are listed in the supplementary material (Tables S2 and S3). View Large Linearized plasmids containing the respective gene fragments were prepared as qPCR standards; in brief, PCR reaction mixtures contained 12.5 µl HotStar Taq Master Mix (Qiagen), 0.2 µg µl−1 Bovine Serum Albumin (BSA), 0.2 pmol µl−1 of each primer, and 1 µl of template DNA per 25 µl reaction. Initial denaturation was at 95°C for 15 min, followed by 30 cycles of 94°C for 60 s, 30 s of annealing (for temperatures see Tables 1 and S4) and elongation at 72°C for 30 s; the final elongation was 10 min at 72°C. Annealing temperatures were determined by temperature gradient PCR, which ranged from 5°Cabove to 5°Cbelow the suggested Tm for the newly designed primers. PCR products were purified (GenElute PCR Clean-Up kit; Sigma) and cloned into E. coli JM 109 using the pGEM-T vector system (Promega); plasmids were purified (GenElute Plasmid Miniprep kit; Sigma) and the insert sequence verified by Sanger sequencing (Macrogen); finally, the plasmids were linearized by restriction digestion with ApaI (Promega) at 37°C for 3.5 h, and quantified by Qubit dsDNA HS Assay kit (Thermo Fisher Scientific). The copy number of the target gene was calculated by the equation \begin{eqnarray*} &&{copy\,number ({{\mu}l}^{-1})}\\ &=& \frac{{[6.023\!\times\!{{10}^{23}}molecules\,mo{l^{ - 1}}\!\times\!Conc \, {plasmid}\,\,(g\,{\rm{\mu }}{l^{ - 1}})]}}{{[fragment\,size\,(bp) \!\times\!660\,g\,mo{l^{ - 1}}\,\!bp{^{ - 1}}]}} \end{eqnarray*} A dilution series with 101– 107 copies µl−1 was used as standard for qPCR. qPCR qPCR reaction mixtures contained 10 µl SYBR Green I Master (Roche), 2 µg µl−1 BSA, and 0.35 pmol µl−1 of each primer, as well as 2 µl of template DNA per 20 µl reaction. qPCR conditions for initial denaturation were 95°C for 5 min, followed by 40 cycles of denaturation (95°C, 30 s), 20 s of primer annealing (see Tables 1 and S4 for temperature), elongation (72°C, 10 s), and fluorescence signal detection for 10 s (see Tables 1 and S4 for temperature). Melting curve analysis (temperature gradient at 1°C steps from 55°C to 95°C) and gel electrophoresis (2% agarose gels) confirmed the specificity of the amplification. qPCR efficiencies for standard curves in all tests were between 86–109%. Expression data are displayed as transcript copies per target gene copy for each of the three terminal oxidase genes. The change of transcription levels over time is expressed as fold change relative to the first sampling point (Pfaffl 2001) after normalization of transcript numbers with the rpoB reference transcript numbers. Normalization with 16S rRNA copy numbers or target gene copy numbers yielded similar results (Figs S3 and S4). Data were log-transformed, and a t test was used to compare all time points with the first data point of the incubation for significant differences. RESULTS AND DISCUSSION Bacterial physiology under low O2 concentration Transcription of terminal oxidases was followed by transition of triplicate cultures from 10 µmol l−1 to <200 nmol l−1 O2, and then for ~ 3 h at <200 nmol l−1 O2. An important consideration for interpretation of the transcription data was to assume that O2 was the only limiting substrate, and thus likely regulating the gene expression. This was attempted by diluting an exponentially growing overnight culture into fresh marine broth media. A relatively stable population size (Fig. S5) and slightly increased bulk respiration rate (Fig. S6) throughout the incubation meant that the respiration rate per cell did not decrease, which indicates that electron donors were not limiting. This non-limiting electron-donor condition (Gong et al. 2016) ensured that O2 concentrations were responsible for the change of gene expression during the incubation. The initial O2 concentration was ~ 10 µmol l−1 at the beginning of the incubation, but the bacteria consumed the O2 and reduced it to nanomolar levels within 2 h (Fig. 1A, B). During the period under low O2 concentrations, the O2 concentrations were kept < 200 nmol l−1, except for short periods when we took samples or adjusted the O2 concentration in the system resulting in slight O2 contamination. Figure 1. View largeDownload slide Time course of oxygen concentration during the incubation for over 300 min for (A) I. loihiensis and (B) M. daepoensis, and relative expression of genes of high-affinity and low-affinity terminal oxidases in (C, E, G) I. loihiensis and (D, F, H) M. daepoensis during the incubation, normalized by transcripts of the rpoB reference gene. The black horizontal lines represent the 200 nmol l−1 oxygen concentration in (A) and (B), and represent the starting gene expression level at an oxygen concentration of 10 µmol l−1 in (C)-(H). Red, green, and black symbols represent three biological replicates. Time 0 was defined as when the oxygen concentration reached 200 nmol l−1 during the incubation. The samples were taken at oxygen concentrations of 10 µmol l−1, 5 µmol l−1, 200 nmol l−1 (time 0), and then 5, 10, 30, 60, 120 and 200 min after the oxygen concentration was below 200 nmol l−1. Figure 1. View largeDownload slide Time course of oxygen concentration during the incubation for over 300 min for (A) I. loihiensis and (B) M. daepoensis, and relative expression of genes of high-affinity and low-affinity terminal oxidases in (C, E, G) I. loihiensis and (D, F, H) M. daepoensis during the incubation, normalized by transcripts of the rpoB reference gene. The black horizontal lines represent the 200 nmol l−1 oxygen concentration in (A) and (B), and represent the starting gene expression level at an oxygen concentration of 10 µmol l−1 in (C)-(H). Red, green, and black symbols represent three biological replicates. Time 0 was defined as when the oxygen concentration reached 200 nmol l−1 during the incubation. The samples were taken at oxygen concentrations of 10 µmol l−1, 5 µmol l−1, 200 nmol l−1 (time 0), and then 5, 10, 30, 60, 120 and 200 min after the oxygen concentration was below 200 nmol l−1. The apparent Km values for I. loihiensis and M. daepoensis were below 100 nmol l−1 throughout the incubation (Fig. S7), which suggests that high-affinity terminal oxidases were expressed. Apparent Km values below 60 nmol l−1 were observed throughout a previous study when these two species were transferred from high to low O2 concentrations (Gong et al. 2016). The samples for tracking the change of transcripts in this study were therefore taken frequently and immediately after injecting bacteria for incubation to resolve a potentially fast change in gene expression after the transfer. The absolute abundances of transcripts for the three types of terminal oxidases were higher in M. daepoensis than in I. loihiensis (Fig. S8), which might be correlated with the fact that the cell size of M. daepoensis (Yoon et al. 2004) was larger than that of I. loihiensis (Donachie 2003) and, according to our previous study (Gong et al. 2016), the non-limited respiration rate per cell of M. daepoensis (2.61 ± 0.21 fmol O2 cell−1 h−1) was also higher than that of I. loihiensis (0.85 ± 0.05 fmol O2 cell−1 h−1). Transcription of the high-affinity cytochrome cbb3 terminal oxidase The abundance of transcripts of the high-affinity cytochrome cbb3 terminal oxidase (C-class HCO) (fixN gene) was highest among the three types of terminal oxidases in both strains, even under the high O2 conditions (Fig. 1, Fig. S8). The observed low apparent Km values (below 60 nmol l−1) measured in the first oxygen depletion curves of the present study, and also in the previous study (Gong et al. 2016), suggest that this cbb3-type oxidase plays an important role for aerobic respiration even under high O2 conditions. Similar to these findings, previous studies in P. aeruginosa also proved the importance of cytochrome cbb3 terminal oxidase under oxic and micro-oxic conditions, even being the dominant oxidase at high O2 conditions (Comolli and Donohue 2004; Alvarez-Ortega and Harwood 2007; Arai et al. 2014). The abundance of fixN transcripts in I. loihiensis (Fig. S8A) was lower than in M. daepoensis (Fig. S8B), i.e. the transcript number could be more than 500 copies per gene in M. daepoensis (Fig. S8B), while it reached only a maximum of 30 copies per gene in I. loihiensis (Fig. S8A). In both strains, the fixN showed by far the clearest up-regulation upon transition to low O2. Irrespective of the normalization method (with rpoB transcript number, Fig. 1; with 16S rRNA transcript number, Fig. S3; or with target gene copy number in DNA, Fig. S4), the fixN transcripts showed similar regulation patterns in I. loihiensis (Fig. 1C) and M. daepoensis (Fig. 1D). fixN was up-regulated significantly in I. loihiensis (Fig. 1C) when samples were exposed to low O2 concentration for 10 min (P < 0.05). The transcripts in M. daepoensis (Fig. 1D) were up-regulated significantly after being exposed to low O2 concentration for 200 min (P < 0.05). The much higher abundance of transcripts of cytochrome cbb3 terminal oxidase than the transcripts of the other two types of terminal oxidases, and the up-regulation of expression under low-O2 concentration conditions, indicates that C-class HCOs are the main terminal oxidases used under micro-oxic conditions in the two investigated strains. The C-class HCOs, mainly confined to the proteobacteria, represent a distinct class of HCOs (Sousa et al. 2012). The more narrow distribution of C-class HCOs, contrasting with the universal distribution of other HCOs within all domains of life, their high affinity for O2, the unique cbb3 central subunit architecture (Buschmann et al. 2010), and the distinctly different subunit composition compared with other HCOs (Garcia-Horsman et al. 1994), suggest that C-class HCOs might be recently evolved enzymes specialized for low O2-condition metabolism (Pereira, Santana and Teixeira 2001). It should also be noted that there are significant differences in the electron donors utilized by HCOs (Morris and Schmidt 2013): C-class HCOs appear to utilize only cytochrome c, not quinol, as electron donors (Keefe and Maier, 1993), while cytochrome bd-type oxidases are exclusive quinol oxidases, using mostly ubiquinol or menaquinol as substrates (Borisov et al. 2011). The differences in electron donors for terminal oxidases could contribute to different efficiencies of energy conservation, which might explain the difference in the abundance of transcripts of these three types of terminal oxidases during the incubation. Transcription of the low-affinity cytochrome aa3 terminal oxidase The abundance of transcripts of the low-affinity cytochrome aa3 terminal oxidase (A-class HCO) (ctaD gene) was low in both I. loihiensis (Fig. S8C) and M. daepoensis (Fig. S8D). The transcript number was lower than one copy per gene, even in the first sample, which contained ~ 10 µmol l−1 O2. Low-affinity terminal oxidases are important during aerobic growth due to a high energy conversion efficiency (Stam et al. 1984; Pronk et al. 1995). Considering that the bacteria grew aerobically, the required amount of low-affinity terminal oxidase might have been synthesized before we took the sample, and the detected low transcript numbers do not therefore necessarily represent low enzymatic activity. Continuous expression of low-affinity cytochrome aa3 oxidase even after exposure to low O2 concentrations for longer periods (Figs S8C, S8D) is consistent with a previous study, where cytochrome aa3 oxidase was expressed in the anaerobe D. vulgaris during anaerobic growth, although the high-affinity terminal oxidase was much higher expressed than the low-affinity cytochrome aa3 oxidase (Lamrabet et al. 2011). In our study, transcription of ctaD was actually up-regulated in both I. loihiensis (Fig. 1E) and M. daepoensis (Fig. 1F) upon transition to low O2 concentration, and remained at higher expression levels throughout the low O2 phase. Although transcript levels remained low, hardly reaching one transcript per gene (Figs S8C, S8D), the up-regulation was about 2- to 5-fold and significant (P < 0.05). A study of a Paracoccus denitrificans chemostat culture (Bosma et al. 1987) showed that the concentration of low-affinity terminal oxidase under fully oxic conditions was much larger than under O2-limited and anoxic conditions, which is in contrast to the up-regulation of low-affinity terminal oxidase gene expression under low O2-concentration conditions observed in this study. Overall, the regulation of terminal oxidase gene expression appears to be highly variable between organisms. The C-class HCOs have high affinity for O2, but their efficiency of energy conservation is lower than for the A-class HCOs. The stoichiometry of proton pumping is 1 H+/e− in A-class HCOs, whereas it is around 0.5 H+/e− in C-class HCOs (Han et al. 2011). Thus, the A-class HCOs require less O2, which is a limiting substrate under low O2-concentration conditions, than C-class HCOs, to generate the equivalent membrane potential (Han et al. 2011). Seen from the perspective of the individual bacterium, the competition is, however, not on resource efficiency but on fastest growth under any given condition, which in this case means the highest amount of proton pumping at any given O2 concentration. An optimized balance between electron flow through the low- and high-affinity terminal oxidases with given Km and maximum rate (Vmax) values will thus be determined by the diffusional supply of O2 to the population of terminal oxidases in the individual cell (Morris and Schmidt 2013). Electron flow through the low-affinity oxidases is preferential until O2 limitation due to kinetic parameters, amounts of terminal oxidases, and diffusion, results in a lowered electron flow that overall produces less proton export than by alternative flow with high-affinity terminal oxidases. The energetic advantage of A-class HCOs might explain the up-regulation of low-affinity terminal oxidase when the samples were exposed to low O2-concentration conditions. The low O2 conditions, however, also induced the up-regulation of expression of high-affinity terminal oxidase (cytochrome cbb3 terminal oxidase). Transcription of the putative low-affinity cytochrome bd-type oxidase, CIO The cytochrome bd-type oxidases are noteworthy for their high affinity for O2, especially the cytochrome bd-I terminal oxidase in E. coli, and resistance to inhibition by cyanide. They are essential for O2 consumption under micro-oxic conditions for bacteria such as E. coli (Cotter et al. 1990), D. vulgaris (Lamrabet et al. 2011), and Bacteroides fragilis (Baughn and Malamy, 2004). Further studies revealed a sub-group of quinol oxidases with a short Q-loop in subunit I and significantly higher resistance to cyanide than the initially classified classical cytochrome bd oxidases, named as cyanide-insensitive oxidases (CIO). CIOs are phylogenetically distinct from classical cytochrome bd oxidases (Miura et al. 2013). Based on recent oxygen affinity data and CIO phylogeny (Fig. S1), the CIO clades with Gluconobacter and Pseudomonas have low affinity for oxygen. Since the I. loihiensis CIO groups with Gluconobacter (Miura et al. 2013), and the Marinobacter CIO, although partially separated from Gluconobacter, is still most closely affiliated with this clade but far from other short Q-loop cytochrome bd oxidases (Fig. S1), we expected the CIOs in the two investigated strains to have low affinity for oxygen. The abundance of transcripts of these CIOs (cydA gene) was generally of the same order as the other low-affinity (cytochrome c) terminal oxidase in I. loihiensis (Figs S8C and S8E), while in M. daepoensis, transcripts of CIO were about 100 times more abundant than transcripts of cytochrome aa3 terminal oxidase (Figs S8D and S8F). This also means that, compared to I. loihiensis (Fig. S8E), the cydA gene was more highly expressed in M. daepoensis (Fig. S8F). The expression of the cydA gene in I. loihiensis (Fig. 1G) was significantly up-regulated (P < 0.05) after being exposed to low O2-concentration conditions (5 min, or even less). After prolonged low O2 concentration, the expression of cydA gene was down-regulated, and had the tendency to return to near-oxic levels, although the expression level was still significantly higher (P < 0.05) at the end of the incubation than under the initial oxic state. Interestingly, the expression of cydA was repressed significantly (P < 0.05) in M. daepoensis already at an O2 concentration of 5 µmol l−1 and after transition to low O2 conditions (Fig. 1H), with the exception of one of the three biological replicates (Fig. 1H: green dots); however, even in this case, cydA gene expression was down-regulated after 30 min under low O2 concentration. The repressed expression of CIO in M. daepoensis, and the fast down-regulation to the level of the fully oxic state after a brief up-regulation upon transition to low O2 conditions during the incubation of I. loihiensis, indicate that CIO expression may not be exclusively regulated by O2. As suggested for other bacterial species, CIO might have other crucial roles in physiology other than terminal oxidation. The characteristics of low affinity for O2 and low efficiency of energy conservation suggest that CIOs may act as a complementary enzyme when the HCOs are not functioning or are inhibited under stressful conditions (Arai et al. 2014). More generally, the up-regulation of all three types of terminal oxidases, regardless of their oxygen affinity, under nanomolar oxygen concentrations, could be a general stress response of the investigated strains because oxygen was the only electron acceptor in the system. The lessons learnt from the gene expression of terminal oxidases in pure cultures Although we can now describe the regulation of terminal oxidase transcription in response to nanomolar O2 concentrations, it is still not clear how the cells maintain a high respiratory rate when exposed to low O2 concentrations (Fig. S6): Do they replace the low-affinity with high-affinity terminal oxidases, do they increase the cellular concentration of high-affinity terminal oxidase without changing that of low-affinity terminal oxidases, or do they apply a combination of these two strategies? For example, in E. coli, both high-affinity terminal oxidases and low-affinity terminal oxidases are present in cells from oxygen-limited continuous cultures (Rice and Hempfling 1978). This lack of knowledge presents an additional limitation when interpreting environmental transcript data. Only one set of metatranscriptomic data from OMZs (off Chile and Peru) reported the distribution of genes and transcripts of terminal oxidases. The study showed that the abundances of transcripts for terminal oxidases were low, that the low-affinity terminal oxidase was expressed in all the sampling layers including the OMZ cores, and that the ratio of high-affinity terminal oxidases to low-affinity terminal oxidases was highest in samples from anoxic layers in terms of both genes and transcripts. However, the transcripts for low-affinity terminal oxidases were present in much higher numbers than those for high-affinity terminal oxidases (Kalvelage et al. 2015). The high number of transcripts for low-affinity terminal oxidases in OMZ samples may be due to the lack of genes for high-affinity terminal oxidases in many marine prokaryotes (Morris and Schmidt 2013), whereas the two strains investigated in our study have both types of genes and thus the ability to adapt to a low O2 environment by modulating their gene expression. This study shows that the pattern of expression of terminal oxidase genes is species-dependent, which illustrates the difficulty of connecting respiratory activities with gene expression, even in pure cultures, let alone in the environment. More detailed knowledge at combined transcript, proteome and activity levels is thus needed for meaningful interpretation of in situ transcriptome data in environmental samples (e.g. OMZs). SUPPLEMENTARY DATA Supplementary data are available at FEMSEC online. ACKNOWLEDGEMENTS We thank Susanne Nielsen, Preben G. Sørensen, Lars B. Pedersen, Trine Bech Søgaard, Britta Poulsen, and Anne Stentebjerg at Aarhus University for their technical assistance, Ian P. G. Marshall at Aarhus University for helpful discussions, and Minenosuke Matsutani and Kazunobu Matsushita at Yamaguchi University for the phylogenetic analysis. FUNDING This work was supported by the European Research Council [grant number 267233]. The funding agency had no role in study design, data collection and interpretation, or the decision to submit the work for publication. Conflicts of interest. None declared. REFERENCES Alvarez-Ortega C , Harwood CS . Responses of Pseudomonas aeruginosa to low oxygen indicate that growth in the cystic fibrosis lung is by aerobic respiration . Mol Microbiol . 2007 ; 65 : 153 – 65 . Google Scholar CrossRef Search ADS PubMed Arai H . Regulation and function of versatile aerobic and anaerobic respiratory metabolism in Pseudomonas aeruginosa . Front Microbiol . 2011 ; 2 : 103 . Google Scholar CrossRef Search ADS PubMed Arai H , Kawakami T , Osamura T et al. Enzymatic characterization and in vivo function of five terminal oxidases in Pseudomonas aeruginosa . J Bacteriol . 2014 ; 196 : 4206 – 15 . Google Scholar CrossRef Search ADS PubMed Baughn AD , Malamy MH . The strict anaerobe Bacteroides fragilis grows in and benefits from nanomolar concentrations of oxygen . Nature . 2004 ; 427 : 441 – 4 . Google Scholar CrossRef Search ADS PubMed Borisov SM , Lehner P , Klimant I . Novel optical trace oxygen sensors based on platinum(II) and palladium(II) complexes with 5,10,15,20-meso-tetrakis-(2,3,4,5,6-pentafluorphenyl)-porphyrin covalently immobilized on silica-gel particles . Anal Chim Acta . 2011 ; 690 : 108 – 15 . Google Scholar CrossRef Search ADS PubMed Borisov VB , Gennis RB , Hemp J et al. The cytochrome bd respiratory oxygen reductases . Biochim Biophys Acta - Bioenerg . 2011 ; 1807 : 1398 – 413 . Google Scholar CrossRef Search ADS Bosma G , Braster M , Stouthamer AH et al. Isolation and characterization of ubiquinol oxidase complexes from Paracoccus denitrificans cells cultured under various limiting growth conditions in the chemostat . Eur J Biochem . 1987 ; 165 : 657 – 63 . Google Scholar CrossRef Search ADS PubMed Brändén G , Gennis RB , Brzezinski P . Transmembrane proton translocation by cytochrome c oxidase . Biochim Biophys Acta . 2006 ; 1757 : 1052 – 63 . Google Scholar CrossRef Search ADS PubMed Buschmann S , Warkentin E , Xie H et al. The structure of cbb 3 cytochrome oxidase provides insights into proton pumping . Science . 2010 ; 329 : 327 – 30 . Google Scholar CrossRef Search ADS PubMed Calhoun MW , Thomas JW , Gennis RB . The cytochrome oxidase superfamily of redox-driven proton pumps . Trends Biochem Sci . 1994 ; 19 : 325 – 30 . Google Scholar CrossRef Search ADS PubMed Canfield DE , Kristensen E , Thamdrup B . Aquatic Geomicrobiology . San Diego : Elsevier Academic Press , 2005 . Comolli JC , Donohue TJ . Differences in two Pseudomonas aeruginosa cbb 3 cytochrome oxidases . Mol Microbiol . 2004 ; 51 : 1193 – 203 . Google Scholar CrossRef Search ADS PubMed Cotter PA , Chepuri V , Gennis RB et al. Cytochrome o (cyoABCDE) and d (cydAB) oxidase gene expression in Escherichia coli is regulated by oxygen, pH, and the fnr gene product . J Bacteriol . 1990 ; 172 : 6333 – 8 . Google Scholar CrossRef Search ADS PubMed D'mello R , Hill S , Poole RK . The cytochrome bd quinol oxidase in Escherichia coli has an extremely high oxygen affinity and two oxygen-binding haems: implications for regulation of activity in vivo by oxygen inhibition . Microbiology . 1996 ; 142 : 755 – 63 . Google Scholar CrossRef Search ADS PubMed Donachie SP . Idiomarina loihiensis sp. nov., a halophilic -Proteobacterium from the Lo'ihi submarine volcano, Hawai'i . Int J Syst Evol Microbiol . 2003 ; 53 : 1873 – 9 . Google Scholar CrossRef Search ADS PubMed Garcia-Horsman JA , Berry E , Shapleigh JP et al. A novel cytochrome c oxidase from Rhodobacter sphaeroides that lacks CuA . Biochemistry . 1994 ; 33 : 3113 – 9 . Google Scholar CrossRef Search ADS PubMed Garcia-Robledo E , Padilla CC , Aldunate M et al. Cryptic oxygen cycling in anoxic marine zones . Proc Natl Acad Sci USA . 2017 ; 114 : 8319 – 24 . Google Scholar CrossRef Search ADS PubMed Gong X , Garcia-Robledo E , Schramm A et al. Respiratory kinetics of marine bacteria exposed to decreasing oxygen concentrations . Appl Environ Microbiol . 2016 ; 82 : 1412 – 22 . Google Scholar CrossRef Search ADS Han H , Hemp J , Pace LA et al. Adaptation of aerobic respiration to low O2 environments . Proc Natl Acad Sci USA . 2011 ; 108 : 14109 – 14 . Google Scholar CrossRef Search ADS PubMed Helly JJ , Levin LA . Global distribution of naturally occurring marine hypoxia on continental margins . Deep Sea Res Part I Oceanogr Res Pap . 2004 ; 51 : 1159 – 68 . Google Scholar CrossRef Search ADS Hemp J , Robinson DE , Ganesan KB et al. Evolutionary migration of a post-translationally modified active-site residue in the proton-pumping heme-copper oxygen reductases . Biochemistry . 2006 ; 45 : 15405 – 10 . Google Scholar CrossRef Search ADS PubMed Jackson RJ , Elvers KT , Lee LJ et al. Oxygen reactivity of both respiratory oxidases in Campylobacter jejuni: the cydAB genes encode a cyanide-resistant, low-affinity oxidase that is not of the cytochrome bd type . J Bacteriol . 2007 ; 189 : 1604 – 15 . Google Scholar CrossRef Search ADS PubMed Kalvelage T , Lavik G , Jensen MM et al. Aerobic microbial respiration in oceanic oxygen minimum zones . PLoS One . 2015 ; 10 : e0133526 . Google Scholar CrossRef Search ADS PubMed Keefe RG , Maier RJ . Purification and characterization of an O2-utilizing cytochrome-c oxidase complex from Bradyrhizobium japonicum bacteroid membranes . Biochim Biophys Acta - Bioenerg . 1993 ; 1183 : 91 – 104 . Google Scholar CrossRef Search ADS Lamrabet O , Pieulle L , Aubert C et al. Oxygen reduction in the strict anaerobe Desulfovibrio vulgaris Hildenborough: characterization of two membrane-bound oxygen reductases . Microbiology . 2011 ; 157 : 2720 – 32 . Google Scholar CrossRef Search ADS PubMed Lehner P , Larndorfer C , Garcia-Robledo E et al. LUMOS - A sensitive and reliable optode system for measuring dissolved oxygen in the nanomolar range . PLoS One . 2015 ; 10 : e0128125 . Google Scholar CrossRef Search ADS PubMed Lever MA , Torti A , Eickenbusch P et al. A modular method for the extraction of DNA and RNA, and the separation of DNA pools from diverse environmental sample types . Front Microbiol . 2015 ; 6 : 476 . Google Scholar CrossRef Search ADS PubMed Miura H , Mogi T , Ano Y et al. Cyanide-insensitive Quinol Oxidase (CIO) from Gluconobacter oxydans is a unique terminal oxidase subfamily of cytochrome bd . J Biochem . 2013 ; 153 : 535 – 45 . Google Scholar CrossRef Search ADS PubMed Morris RL , Schmidt TM . Shallow breathing: bacterial life at low O2 . Nat Rev Microbiol . 2013 ; 11 : 205 – 12 . Google Scholar CrossRef Search ADS PubMed Pereira MM , Santana M , Teixeira M . A novel scenario for the evolution of haem–copper oxygen reductases . Biochim Biophys Acta - Bioenerg . 2001 ; 1505 : 185 – 208 . Google Scholar CrossRef Search ADS Pfaffl MW . A new mathematical model for relative quantification in real-time RT-PCR . Nucleic Acids Res . 2001 ; 29 : e45 . Google Scholar CrossRef Search ADS PubMed Pitcher RS , Watmough NJ . The bacterial cytochrome cbb 3 oxidases . Biochim Biophys Acta - Bioenerg . 2004 ; 1655 : 388 – 99 . Google Scholar CrossRef Search ADS Poole RK . Bacterial cytochrome oxidases: A structurally and functionally diverse group of electron-transfer protein . Biochim Biophys Acta - Rev Bioenerg . 1983 ; 726 : 205 – 43 . Google Scholar CrossRef Search ADS Pronk AF , Affourtit C , Mashingaidze C et al. The relationship between the in situ reduction level of the cytochrome c pool of Azorhizobium caulinodans growing in a chemostat with NH4+ or N2 as the N source and the total activity of cytochrome c oxidases . FEMS Microbiol Lett . 1995 ; 129 : 149 – 55 . Revsbech NP , Larsen LH , Gundersen J et al. Determination of ultra-low oxygen concentrations in oxygen minimum zones by the STOX sensor . Limnol Ocean Methods . 2009 ; 7 : 371 – 81 . Google Scholar CrossRef Search ADS Rice CW , Hempfling WP . Oxygen-limited continuous culture and respiratory energy conservation in Escherichia coli . J Bacteriol . 1978 ; 134 : 115 – 24 . Google Scholar PubMed Stam H , van Verseveld HW , de Vries W et al. Hydrogen oxidation and efficiency of nitrogen fixation in succinate-limited chemostat cultures of Rhizobium ORS 571 . Arch Microbiol . 1984 ; 139 : 53 – 60 . Google Scholar CrossRef Search ADS Stolper DA , Revsbech NP , Canfield DE . Aerobic growth at nanomolar oxygen concentrations . Proc Natl Acad Sci USA . 2010 ; 107 : 18755 – 60 . Google Scholar CrossRef Search ADS PubMed Sousa FL , Alves RJ , Ribeiro MA et al. The superfamily of heme–copper oxygen reductases: Types and evolutionary considerations . Biochim Biophys Acta - Bioenerg . 2012 ; 1817 : 629 – 37 . Google Scholar CrossRef Search ADS Wright JJ , Konwar KM , Hallam SJ . Microbial ecology of expanding oxygen minimum zones . Nat Rev Microbiol . 2012 ; 10 : 381 – 94 . Google Scholar CrossRef Search ADS PubMed Yoon J-H , Yeo S-H , Kim I-G et al. Marinobacter flavimaris sp. nov. and Marinobacter daepoensis sp. nov., slightly halophilic organisms isolated from sea water of the Yellow Sea in Korea . Int J Syst Evol Microbiol . 2004 ; 54 : 1799 – 803 . Google Scholar CrossRef Search ADS PubMed © FEMS 2018. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png FEMS Microbiology Ecology Oxford University Press

Gene expression of terminal oxidases in two marine bacterial strains exposed to nanomolar oxygen concentrations

Loading next page...
 
/lp/ou_press/gene-expression-of-terminal-oxidases-in-two-marine-bacterial-strains-26huK6TtuA
Publisher
Blackwell
Copyright
© FEMS 2018.
ISSN
0168-6496
eISSN
1574-6941
D.O.I.
10.1093/femsec/fiy072
Publisher site
See Article on Publisher Site

Abstract

Abstract The final step of aerobic respiration is carried out by a terminal oxidase transporting electrons to oxygen (O2). Prokaryotes harbor diverse terminal oxidases that differ in phylogenetic origin, structure, biochemical function, and affinity for O2. Here we report on the expression of high-affinity (cytochrome cbb3 oxidase), low-affinity (cytochrome aa3 oxidase), and putative low-affinity (cyanide-insensitive oxidase (CIO)) terminal oxidases in the marine bacteria Idiomarina loihiensis L2-TR and Marinobacter daepoensis SW-156 upon transition to very low O2 concentrations (<200 nM), measured by RT-qPCR. In both strains, high-affinity cytochrome cbb3 oxidase showed the highest expression levels and was significantly up-regulated upon transition to low O2 concentrations. Low-affinity cytochrome aa3 oxidase showed very low transcription levels throughout the incubation. Surprisingly, however, it was also up-regulated upon transition to low O2 concentrations. In contrast, putative low-affinity CIO had much lower expression levels and markedly different regulation patterns between the two strains. These results demonstrate that exposure to low O2 concentrations regulates the gene expression of different types of terminal oxidases, but also that the type and magnitude of transcriptional response is species-dependent. Therefore, in situ transcriptome data cannot, without detailed knowledge of the transcriptional regulation of the species involved, be translated into relative respiratory activity. nanomolar oxygen, gene expression, terminal oxidase, RT-qPCR, marine bacteria INTRODUCTION Oxygen (O2) is the most favourable electron acceptor in microbial respiration (Canfield, Kristensen and Thamdrup 2005). It may therefore not seem surprising that O2 can be used as a terminal electron acceptor even at nanomolar concentrations (Stolper, Revsbech and Canfield 2010). This is ecologically relevant in many micro-oxic environments, e.g. in marine oxygen minimum zones (OMZs). OMZs are oceanic areas with a layer of seawater at intermediate depth with an O2 concentration <20 µmol l−1 (Helly and Levin 2004), sometimes fluctuating from a few micromolar to as low as a few nanomolar. OMZs have a global impact on the biogeochemistry of the oceans, especially with respect to nitrogen cycling (Wright, Konwar and Hallam 2012), and show high rates of aerobic respiration despite low O2 concentrations (Garcia-Robledo et al. 2017). The biochemical basis for this are terminal oxidases with very high O2 affinities, i.e. with half-saturation constants (Km) in the range of 3–8 nmol l−1 (D'mello, Hill and Poole 1996). Terminal oxidases transfer electrons to O2 in the final step of aerobic respiration (Poole 1983), and have been classified into two main families according to their structural and functional differences (Calhoun, Thomas and Gennis 1994; Pitcher and Watmough 2004; Brändén, Gennis and Brzezinski 2006; Hemp et al. 2006): (i) the heme-copper oxidases (HCOs), which can be subdivided into A-class HCOs (with low O2 affinity) and B- and C-class HCOs (with high O2 affinity) (Morris and Schmidt 2013); and ii) the cytochrome bd quinol oxidases, which can be subdivided into classical bd-type (with high O2 affinity) (Borisov et al. 2011) and cyanide-insensitive oxidases (CIOs: putative with low O2 affinity) (Jackson et al. 2007). CIOs are homologs of classical cytochrome bd oxidases, but are phylogenetically distinct (Miura et al. 2013). Kinetic analysis showed that CIOs in diverse bacteria, e.g. Pseudomonas aeruginosa (Arai et al. 2014), Campylobacter jejuni (Jackson et al. 2007) and Gluconobacter oxydans (Miura et al. 2013), have low affinity for O2, even comparable to low-affinity A-class HCOs. O2 concentration has been shown to regulate terminal oxidase expression in a few bacterial species containing both high- and low-affinity enzymes (e.g. P. aeruginosa (Alvarez-Ortega and Harwood 2007; Arai 2011; Arai et al. 2014), Desulfovibrio vulgaris (Lamrabet et al. 2011), and Escherichia coli (Cotter et al. 1990)), albeit never under nanomolar O2 concentrations. Environmental data suggest that genes for both high- and low-affinity terminal oxidases are widely distributed and simultaneously expressed in diverse environments (Morris and Schmidt 2013), including OMZs under varying O2 concentrations (Kalvelage et al. 2015). The link between O2 concentration, terminal oxidase expression, and the kinetics of O2 respiration is thus not well understood, in particular not at nanomolar O2 concentrations. Only recently, novel sensor developments (Revsbech et al. 2009; Lehner et al. 2015) have enabled detailed measurements of O2 respiration rates and kinetics at nanomolar O2 concentrations, both in situ (Garcia-Robledo et al. 2017) and in controlled pure culture incubations (Gong et al. 2016). The latter study followed the change in O2 respiration kinetics of four marine bacterial strains during the transition to nanomolar O2 concentrations and revealed apparent Km values of 10–60 nmol l−1 (Gong et al. 2016). The aim of the present study was to follow up on this kinetic study (Gong et al. 2016) by quantifying the expression of both low- and high-affinity terminal oxidases during the transition from aerobic to nearly anoxic conditions, in order to link gene expression with respiration kinetics. To achieve this we selected two of the investigated strains (Idiomarina loihiensis L2-TR DSM-15 497 and Marinobacter daepoensis SW-156 DSM-16 072) that (i) carry single-copy genes of high-affinity terminal oxidase (cytochrome cbb3 oxidase, C-class HCO), low-affinity terminal oxidase (cytochrome aa3 oxidase, A-class HCO) and putative low-affinity terminal oxidase (cytochrome bd quinol oxidase, CIO; Fig. S1); and (ii) were previously detected in aquatic environments (Fig. S2), including OMZs (Table S1). Terminal oxidase genes and transcripts were detected by reverse transcription quantitative PCR (RT-qPCR) assays specifically developed for all individual genes and strains. MATERIALS AND METHODS Bacterial cultures I. loihiensis L2-TR DSM-15 497 and M. daepoensis SW-156 DSM-16 072 were obtained from the Leibniz Institute DSMZ, Germany. Bacteria were grown at 25°C in autoclaved marine broth media containing 5 g peptone (BD), 1 g yeast extract (Difco), 0.1 g FeC6H5O7, 12.6 g MgCl2·6 H2O, 3.24 g Na2SO4, 19.45 g NaCl, 2.38 g CaCl2·2H2O, 0.55 g KCl, 0.16 g NaHCO3 and 0.01 g Na2HPO4·2H2O l-1. Bacterial cultures were inoculated into fresh marine broth media and incubated overnight on a rotary shaker (180 rpm) to ensure fully oxic conditions and also obtain cell suspensions in the exponential growth phase. Set-up and incubation The highly sensitive optical O2 sensors used in this study have been described previously (Borisov, Lehner and Klimant 2011; Lehner et al. 2015; Gong et al. 2016); in brief, combinations of a fluorophore and immobilizing matrix resulted in two types of O2 optodes: a highly sensitive optode with a measuring range of 1–1000 nmol l−1, and a broader range optode with a measuring range of 0.01–10 µmol l−1. Data were recorded every 10 s during the incubations. Incubation bottles (2 l; Schott) with pre-glued optode dots of both types inside, and a glass-coated magnet in the bottom, were filled with 1.5 l of marine broth medium. Optode dots were positioned 2–3 cm above the bottom to ensure that they were immersed in the liquid phase during the incubation. Two glass tubes went through the butyl rubber bottle cap for flushing the medium with N2 to remove dissolved O2 and to take samples. All bottles, glass-coated magnets and tubes were previously cleaned with 0.1 mol l−1 HCl, and subsequently washed with autoclaved water to minimize bacterial contamination. The optodes and permanently attached optoelectronics (Lehner et al. 2015; Gong et al. 2016) were calibrated before the incubation by addition of various volumes of air-saturated water to freshly deoxygenated water. Incubations were performed in triplicate in a constant-temperature water bath at 25°C during the experiments. Before injecting bacteria, the bottles with media were kept in the water bath for 30 min to reach temperature equilibrium. The incubations lasted for 300–350 min. The O2 concentration was controlled by injecting air-equilibrated artificial seawater or flushing with N2 during the incubation. In order to estimate the growth phase of bacteria in the inoculum, the optical density at 600 nm was measured with a spectrophotometer (Pharmacia Biotech). The value of OD600 of the stock culture was adjusted to 0.5 with marine broth medium before inoculation (Table S2). A volume of 5 ml of the adjusted bacterial inoculum was injected into the medium contained in the bottle when the O2 concentration decreased to 10 µmol l−1 by N2 bubbling. The first sample was taken once the bacteria were injected into the incubation bottle. The bubbling with N2 was terminated after the inoculation, so any further decreases in O2 concentration were due to respiration. The second sample was taken when the O2 concentration had decreased to 5 µmol l−1. Time zero in the subsequent annotation was defined as the time point when the O2 concentration had decreased to 200 nmol l−1. Further samples were taken when the cultures had been exposed to low O2 concentrations (≤ 200 nmol l−1) for 5, 10, 30, 120 and 200 min. Samples of 100 ml were taken by a syringe connected to one of the two glass tubes penetrating the stopper. Samples were filtered onto 0.22-µm pore-size polycarbonate filters (47 mm, Sigma), and immediately transferred to micro-centrifuge tubes containing 250 µl water and 500 µl RNAprotect Bacteria Reagent (Qiagen). All processes from sampling to immersion in RNAprotect Bacteria Reagent were finished within 5 min. The mixture of filter and reagent was vortexed for 5 s and subsequently incubated for 30 min at room temperature. After centrifuging for 10 min at 5000 × g, the supernatant was removed. The samples were stored at −80°C. Nucleic acid extraction and preparation of cDNA Cells on the filter were lysed by the lysozyme and proteinase K method described in the RNeasy Mini kit (Qiagen). Half of the lysate was used for DNA extraction by the DNeasy Blood and Tissue kit (Qiagen) following the manufacturer's protocol. The quality of the DNA extracts was checked by using serial dilutions as templates for qPCR to confirm that the extracts did not contain inhibitors for qPCR. The other half of the cell lysate was used for RNA extraction by the RNeasy Mini kit (Qiagen) according to the manufacturer's instructions. Genomic DNA co-extracted during RNA extraction was removed by TURBO DNA-free kit (Ambion), and the complete removal was confirmed by qPCR of a terminal oxidase gene and a reference gene (see below), both of which were negative for RNA extracts. RNA was checked for integrity on 0.8% agarose electrophoresis gels, quantified by Qubit RNA BR Assay Kit (Thermo Fisher Scientific), and reverse-transcribed using the iScript cDNA Synthesis kit (Bio-rad) following the manufacturer's instructions. The cDNA was stored at −20°C. Primer design, PCR, and qPCR standard The genes of subunit I of each terminal oxidase and of rpoB, encoding the RNA polymerase β subunit (as reference gene), were retrieved from NCBI for I. loihiensis and M. daepoensis and aligned in Geneious 4.8.5. Specific PCR primers (Tables 1 and S3) were designed manually and checked in silico by OligoAnalyzer 3.1 (http://eu.idtdna.com/calc/analyzer). Primers and PCR conditions for the 16S rRNA gene (Tables S3 and S4) were as previously described (Lever et al. 2015). Table 1. Target genes and specific qPCR assays for detecting terminal oxidases in I. loihiensis and M. daepoensis. Species Name I. loihiensis L2-TR DSM-15 497 M. daepoensis SW-156 DSM-16 072 Target protein 1(1) cbb3-type cytochrome c oxidase subunit 1 cbb3-type cytochrome c oxidase subunit 1 Oxidase type (gene) C-class HCO (fixN) C-class HCO (fixN) Oxidase affinity High affinity High affinity Locus tag (source) IL1297 (NCBI) K325DRAFT_1727 (IMG) Primer sequence; name CTGACCGCTGGCTTCCTGGG; Il-fixNF CTGACCGCCGGCTTCCTGGG; Md-fixNF (5′→3′) GTCATGATACCGTTAATCATGCCGCCCCA; Il-fixNR GTCATCATACCGTTGATCATGCCGCCCCA; Md-fixNR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 82 86 Target protein 2 cytochrome aa3 oxidase subunit I cytochrome aa3 oxidase subunit I Oxidase type (gene) A-class HCO (ctaD) A-class HCO (ctaD) Oxidase affinity Low affinity Low affinity Locus tag (source) IL0259 (NCBI) K325DRAFT_2803 (IMG) Primer sequence; name TTCGGTACGAGTTTCTTTGACGCC; Il-ctaDF TTCGGGACAAGTTTCTTTGACGCC; Md-ctaDF (5′→3′) AGTTAAAGATCTTCACCCCGGTGGG; Il-ctaDR AGTTAAACACCTTCACCCCGGTGGG; Md-ctaDR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 83 86 Target protein 3 cytochrome bd-type quinol oxidase subunit 1 cytochrome bd-I ubiquinol oxidase subunit 1 apoprotein Oxidase type (gene) CIO (cydA) CIO (cydA) Oxidase affinity Putative low-affinity Putative low-affinity Locus tag (source) IL0041 (NCBI) K325DRAFT_1300 (IMG) Primer sequence; name GCGGTGGGCACGGCTATTTCAGC; Il-cydAF GCCACGGGTACCTTTATTTCCTC; Md-cydAF (5′→3′) TTTTACGGGCGGCGGCGTTGGT; Il-cydAR CCTTACGGTTTGCCTCTACCTCA; Md-cydAR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 86 86 Species Name I. loihiensis L2-TR DSM-15 497 M. daepoensis SW-156 DSM-16 072 Target protein 1(1) cbb3-type cytochrome c oxidase subunit 1 cbb3-type cytochrome c oxidase subunit 1 Oxidase type (gene) C-class HCO (fixN) C-class HCO (fixN) Oxidase affinity High affinity High affinity Locus tag (source) IL1297 (NCBI) K325DRAFT_1727 (IMG) Primer sequence; name CTGACCGCTGGCTTCCTGGG; Il-fixNF CTGACCGCCGGCTTCCTGGG; Md-fixNF (5′→3′) GTCATGATACCGTTAATCATGCCGCCCCA; Il-fixNR GTCATCATACCGTTGATCATGCCGCCCCA; Md-fixNR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 82 86 Target protein 2 cytochrome aa3 oxidase subunit I cytochrome aa3 oxidase subunit I Oxidase type (gene) A-class HCO (ctaD) A-class HCO (ctaD) Oxidase affinity Low affinity Low affinity Locus tag (source) IL0259 (NCBI) K325DRAFT_2803 (IMG) Primer sequence; name TTCGGTACGAGTTTCTTTGACGCC; Il-ctaDF TTCGGGACAAGTTTCTTTGACGCC; Md-ctaDF (5′→3′) AGTTAAAGATCTTCACCCCGGTGGG; Il-ctaDR AGTTAAACACCTTCACCCCGGTGGG; Md-ctaDR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 83 86 Target protein 3 cytochrome bd-type quinol oxidase subunit 1 cytochrome bd-I ubiquinol oxidase subunit 1 apoprotein Oxidase type (gene) CIO (cydA) CIO (cydA) Oxidase affinity Putative low-affinity Putative low-affinity Locus tag (source) IL0041 (NCBI) K325DRAFT_1300 (IMG) Primer sequence; name GCGGTGGGCACGGCTATTTCAGC; Il-cydAF GCCACGGGTACCTTTATTTCCTC; Md-cydAF (5′→3′) TTTTACGGGCGGCGGCGTTGGT; Il-cydAR CCTTACGGTTTGCCTCTACCTCA; Md-cydAR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 86 86 (1): For ease of use, the same gene symbols coding for the terminal oxidases are used for these two species throughout the study, i.e. target 1: fixN; target 2: ctaD; and target 3: cydA. The rule for this modification was that those genes with different gene symbols coding for the same functional subunit are based on the BLAST annotation and sequence. The detailed primer sequences, annealing temperatures and detection temperatures in qPCR for rpoB and 16S rRNA genes are listed in the supplementary material (Tables S2 and S3). View Large Table 1. Target genes and specific qPCR assays for detecting terminal oxidases in I. loihiensis and M. daepoensis. Species Name I. loihiensis L2-TR DSM-15 497 M. daepoensis SW-156 DSM-16 072 Target protein 1(1) cbb3-type cytochrome c oxidase subunit 1 cbb3-type cytochrome c oxidase subunit 1 Oxidase type (gene) C-class HCO (fixN) C-class HCO (fixN) Oxidase affinity High affinity High affinity Locus tag (source) IL1297 (NCBI) K325DRAFT_1727 (IMG) Primer sequence; name CTGACCGCTGGCTTCCTGGG; Il-fixNF CTGACCGCCGGCTTCCTGGG; Md-fixNF (5′→3′) GTCATGATACCGTTAATCATGCCGCCCCA; Il-fixNR GTCATCATACCGTTGATCATGCCGCCCCA; Md-fixNR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 82 86 Target protein 2 cytochrome aa3 oxidase subunit I cytochrome aa3 oxidase subunit I Oxidase type (gene) A-class HCO (ctaD) A-class HCO (ctaD) Oxidase affinity Low affinity Low affinity Locus tag (source) IL0259 (NCBI) K325DRAFT_2803 (IMG) Primer sequence; name TTCGGTACGAGTTTCTTTGACGCC; Il-ctaDF TTCGGGACAAGTTTCTTTGACGCC; Md-ctaDF (5′→3′) AGTTAAAGATCTTCACCCCGGTGGG; Il-ctaDR AGTTAAACACCTTCACCCCGGTGGG; Md-ctaDR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 83 86 Target protein 3 cytochrome bd-type quinol oxidase subunit 1 cytochrome bd-I ubiquinol oxidase subunit 1 apoprotein Oxidase type (gene) CIO (cydA) CIO (cydA) Oxidase affinity Putative low-affinity Putative low-affinity Locus tag (source) IL0041 (NCBI) K325DRAFT_1300 (IMG) Primer sequence; name GCGGTGGGCACGGCTATTTCAGC; Il-cydAF GCCACGGGTACCTTTATTTCCTC; Md-cydAF (5′→3′) TTTTACGGGCGGCGGCGTTGGT; Il-cydAR CCTTACGGTTTGCCTCTACCTCA; Md-cydAR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 86 86 Species Name I. loihiensis L2-TR DSM-15 497 M. daepoensis SW-156 DSM-16 072 Target protein 1(1) cbb3-type cytochrome c oxidase subunit 1 cbb3-type cytochrome c oxidase subunit 1 Oxidase type (gene) C-class HCO (fixN) C-class HCO (fixN) Oxidase affinity High affinity High affinity Locus tag (source) IL1297 (NCBI) K325DRAFT_1727 (IMG) Primer sequence; name CTGACCGCTGGCTTCCTGGG; Il-fixNF CTGACCGCCGGCTTCCTGGG; Md-fixNF (5′→3′) GTCATGATACCGTTAATCATGCCGCCCCA; Il-fixNR GTCATCATACCGTTGATCATGCCGCCCCA; Md-fixNR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 82 86 Target protein 2 cytochrome aa3 oxidase subunit I cytochrome aa3 oxidase subunit I Oxidase type (gene) A-class HCO (ctaD) A-class HCO (ctaD) Oxidase affinity Low affinity Low affinity Locus tag (source) IL0259 (NCBI) K325DRAFT_2803 (IMG) Primer sequence; name TTCGGTACGAGTTTCTTTGACGCC; Il-ctaDF TTCGGGACAAGTTTCTTTGACGCC; Md-ctaDF (5′→3′) AGTTAAAGATCTTCACCCCGGTGGG; Il-ctaDR AGTTAAACACCTTCACCCCGGTGGG; Md-ctaDR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 83 86 Target protein 3 cytochrome bd-type quinol oxidase subunit 1 cytochrome bd-I ubiquinol oxidase subunit 1 apoprotein Oxidase type (gene) CIO (cydA) CIO (cydA) Oxidase affinity Putative low-affinity Putative low-affinity Locus tag (source) IL0041 (NCBI) K325DRAFT_1300 (IMG) Primer sequence; name GCGGTGGGCACGGCTATTTCAGC; Il-cydAF GCCACGGGTACCTTTATTTCCTC; Md-cydAF (5′→3′) TTTTACGGGCGGCGGCGTTGGT; Il-cydAR CCTTACGGTTTGCCTCTACCTCA; Md-cydAR Annealing temperature (°C) 59 59 Detection temperature in qPCR (°C) 86 86 (1): For ease of use, the same gene symbols coding for the terminal oxidases are used for these two species throughout the study, i.e. target 1: fixN; target 2: ctaD; and target 3: cydA. The rule for this modification was that those genes with different gene symbols coding for the same functional subunit are based on the BLAST annotation and sequence. The detailed primer sequences, annealing temperatures and detection temperatures in qPCR for rpoB and 16S rRNA genes are listed in the supplementary material (Tables S2 and S3). View Large Linearized plasmids containing the respective gene fragments were prepared as qPCR standards; in brief, PCR reaction mixtures contained 12.5 µl HotStar Taq Master Mix (Qiagen), 0.2 µg µl−1 Bovine Serum Albumin (BSA), 0.2 pmol µl−1 of each primer, and 1 µl of template DNA per 25 µl reaction. Initial denaturation was at 95°C for 15 min, followed by 30 cycles of 94°C for 60 s, 30 s of annealing (for temperatures see Tables 1 and S4) and elongation at 72°C for 30 s; the final elongation was 10 min at 72°C. Annealing temperatures were determined by temperature gradient PCR, which ranged from 5°Cabove to 5°Cbelow the suggested Tm for the newly designed primers. PCR products were purified (GenElute PCR Clean-Up kit; Sigma) and cloned into E. coli JM 109 using the pGEM-T vector system (Promega); plasmids were purified (GenElute Plasmid Miniprep kit; Sigma) and the insert sequence verified by Sanger sequencing (Macrogen); finally, the plasmids were linearized by restriction digestion with ApaI (Promega) at 37°C for 3.5 h, and quantified by Qubit dsDNA HS Assay kit (Thermo Fisher Scientific). The copy number of the target gene was calculated by the equation \begin{eqnarray*} &&{copy\,number ({{\mu}l}^{-1})}\\ &=& \frac{{[6.023\!\times\!{{10}^{23}}molecules\,mo{l^{ - 1}}\!\times\!Conc \, {plasmid}\,\,(g\,{\rm{\mu }}{l^{ - 1}})]}}{{[fragment\,size\,(bp) \!\times\!660\,g\,mo{l^{ - 1}}\,\!bp{^{ - 1}}]}} \end{eqnarray*} A dilution series with 101– 107 copies µl−1 was used as standard for qPCR. qPCR qPCR reaction mixtures contained 10 µl SYBR Green I Master (Roche), 2 µg µl−1 BSA, and 0.35 pmol µl−1 of each primer, as well as 2 µl of template DNA per 20 µl reaction. qPCR conditions for initial denaturation were 95°C for 5 min, followed by 40 cycles of denaturation (95°C, 30 s), 20 s of primer annealing (see Tables 1 and S4 for temperature), elongation (72°C, 10 s), and fluorescence signal detection for 10 s (see Tables 1 and S4 for temperature). Melting curve analysis (temperature gradient at 1°C steps from 55°C to 95°C) and gel electrophoresis (2% agarose gels) confirmed the specificity of the amplification. qPCR efficiencies for standard curves in all tests were between 86–109%. Expression data are displayed as transcript copies per target gene copy for each of the three terminal oxidase genes. The change of transcription levels over time is expressed as fold change relative to the first sampling point (Pfaffl 2001) after normalization of transcript numbers with the rpoB reference transcript numbers. Normalization with 16S rRNA copy numbers or target gene copy numbers yielded similar results (Figs S3 and S4). Data were log-transformed, and a t test was used to compare all time points with the first data point of the incubation for significant differences. RESULTS AND DISCUSSION Bacterial physiology under low O2 concentration Transcription of terminal oxidases was followed by transition of triplicate cultures from 10 µmol l−1 to <200 nmol l−1 O2, and then for ~ 3 h at <200 nmol l−1 O2. An important consideration for interpretation of the transcription data was to assume that O2 was the only limiting substrate, and thus likely regulating the gene expression. This was attempted by diluting an exponentially growing overnight culture into fresh marine broth media. A relatively stable population size (Fig. S5) and slightly increased bulk respiration rate (Fig. S6) throughout the incubation meant that the respiration rate per cell did not decrease, which indicates that electron donors were not limiting. This non-limiting electron-donor condition (Gong et al. 2016) ensured that O2 concentrations were responsible for the change of gene expression during the incubation. The initial O2 concentration was ~ 10 µmol l−1 at the beginning of the incubation, but the bacteria consumed the O2 and reduced it to nanomolar levels within 2 h (Fig. 1A, B). During the period under low O2 concentrations, the O2 concentrations were kept < 200 nmol l−1, except for short periods when we took samples or adjusted the O2 concentration in the system resulting in slight O2 contamination. Figure 1. View largeDownload slide Time course of oxygen concentration during the incubation for over 300 min for (A) I. loihiensis and (B) M. daepoensis, and relative expression of genes of high-affinity and low-affinity terminal oxidases in (C, E, G) I. loihiensis and (D, F, H) M. daepoensis during the incubation, normalized by transcripts of the rpoB reference gene. The black horizontal lines represent the 200 nmol l−1 oxygen concentration in (A) and (B), and represent the starting gene expression level at an oxygen concentration of 10 µmol l−1 in (C)-(H). Red, green, and black symbols represent three biological replicates. Time 0 was defined as when the oxygen concentration reached 200 nmol l−1 during the incubation. The samples were taken at oxygen concentrations of 10 µmol l−1, 5 µmol l−1, 200 nmol l−1 (time 0), and then 5, 10, 30, 60, 120 and 200 min after the oxygen concentration was below 200 nmol l−1. Figure 1. View largeDownload slide Time course of oxygen concentration during the incubation for over 300 min for (A) I. loihiensis and (B) M. daepoensis, and relative expression of genes of high-affinity and low-affinity terminal oxidases in (C, E, G) I. loihiensis and (D, F, H) M. daepoensis during the incubation, normalized by transcripts of the rpoB reference gene. The black horizontal lines represent the 200 nmol l−1 oxygen concentration in (A) and (B), and represent the starting gene expression level at an oxygen concentration of 10 µmol l−1 in (C)-(H). Red, green, and black symbols represent three biological replicates. Time 0 was defined as when the oxygen concentration reached 200 nmol l−1 during the incubation. The samples were taken at oxygen concentrations of 10 µmol l−1, 5 µmol l−1, 200 nmol l−1 (time 0), and then 5, 10, 30, 60, 120 and 200 min after the oxygen concentration was below 200 nmol l−1. The apparent Km values for I. loihiensis and M. daepoensis were below 100 nmol l−1 throughout the incubation (Fig. S7), which suggests that high-affinity terminal oxidases were expressed. Apparent Km values below 60 nmol l−1 were observed throughout a previous study when these two species were transferred from high to low O2 concentrations (Gong et al. 2016). The samples for tracking the change of transcripts in this study were therefore taken frequently and immediately after injecting bacteria for incubation to resolve a potentially fast change in gene expression after the transfer. The absolute abundances of transcripts for the three types of terminal oxidases were higher in M. daepoensis than in I. loihiensis (Fig. S8), which might be correlated with the fact that the cell size of M. daepoensis (Yoon et al. 2004) was larger than that of I. loihiensis (Donachie 2003) and, according to our previous study (Gong et al. 2016), the non-limited respiration rate per cell of M. daepoensis (2.61 ± 0.21 fmol O2 cell−1 h−1) was also higher than that of I. loihiensis (0.85 ± 0.05 fmol O2 cell−1 h−1). Transcription of the high-affinity cytochrome cbb3 terminal oxidase The abundance of transcripts of the high-affinity cytochrome cbb3 terminal oxidase (C-class HCO) (fixN gene) was highest among the three types of terminal oxidases in both strains, even under the high O2 conditions (Fig. 1, Fig. S8). The observed low apparent Km values (below 60 nmol l−1) measured in the first oxygen depletion curves of the present study, and also in the previous study (Gong et al. 2016), suggest that this cbb3-type oxidase plays an important role for aerobic respiration even under high O2 conditions. Similar to these findings, previous studies in P. aeruginosa also proved the importance of cytochrome cbb3 terminal oxidase under oxic and micro-oxic conditions, even being the dominant oxidase at high O2 conditions (Comolli and Donohue 2004; Alvarez-Ortega and Harwood 2007; Arai et al. 2014). The abundance of fixN transcripts in I. loihiensis (Fig. S8A) was lower than in M. daepoensis (Fig. S8B), i.e. the transcript number could be more than 500 copies per gene in M. daepoensis (Fig. S8B), while it reached only a maximum of 30 copies per gene in I. loihiensis (Fig. S8A). In both strains, the fixN showed by far the clearest up-regulation upon transition to low O2. Irrespective of the normalization method (with rpoB transcript number, Fig. 1; with 16S rRNA transcript number, Fig. S3; or with target gene copy number in DNA, Fig. S4), the fixN transcripts showed similar regulation patterns in I. loihiensis (Fig. 1C) and M. daepoensis (Fig. 1D). fixN was up-regulated significantly in I. loihiensis (Fig. 1C) when samples were exposed to low O2 concentration for 10 min (P < 0.05). The transcripts in M. daepoensis (Fig. 1D) were up-regulated significantly after being exposed to low O2 concentration for 200 min (P < 0.05). The much higher abundance of transcripts of cytochrome cbb3 terminal oxidase than the transcripts of the other two types of terminal oxidases, and the up-regulation of expression under low-O2 concentration conditions, indicates that C-class HCOs are the main terminal oxidases used under micro-oxic conditions in the two investigated strains. The C-class HCOs, mainly confined to the proteobacteria, represent a distinct class of HCOs (Sousa et al. 2012). The more narrow distribution of C-class HCOs, contrasting with the universal distribution of other HCOs within all domains of life, their high affinity for O2, the unique cbb3 central subunit architecture (Buschmann et al. 2010), and the distinctly different subunit composition compared with other HCOs (Garcia-Horsman et al. 1994), suggest that C-class HCOs might be recently evolved enzymes specialized for low O2-condition metabolism (Pereira, Santana and Teixeira 2001). It should also be noted that there are significant differences in the electron donors utilized by HCOs (Morris and Schmidt 2013): C-class HCOs appear to utilize only cytochrome c, not quinol, as electron donors (Keefe and Maier, 1993), while cytochrome bd-type oxidases are exclusive quinol oxidases, using mostly ubiquinol or menaquinol as substrates (Borisov et al. 2011). The differences in electron donors for terminal oxidases could contribute to different efficiencies of energy conservation, which might explain the difference in the abundance of transcripts of these three types of terminal oxidases during the incubation. Transcription of the low-affinity cytochrome aa3 terminal oxidase The abundance of transcripts of the low-affinity cytochrome aa3 terminal oxidase (A-class HCO) (ctaD gene) was low in both I. loihiensis (Fig. S8C) and M. daepoensis (Fig. S8D). The transcript number was lower than one copy per gene, even in the first sample, which contained ~ 10 µmol l−1 O2. Low-affinity terminal oxidases are important during aerobic growth due to a high energy conversion efficiency (Stam et al. 1984; Pronk et al. 1995). Considering that the bacteria grew aerobically, the required amount of low-affinity terminal oxidase might have been synthesized before we took the sample, and the detected low transcript numbers do not therefore necessarily represent low enzymatic activity. Continuous expression of low-affinity cytochrome aa3 oxidase even after exposure to low O2 concentrations for longer periods (Figs S8C, S8D) is consistent with a previous study, where cytochrome aa3 oxidase was expressed in the anaerobe D. vulgaris during anaerobic growth, although the high-affinity terminal oxidase was much higher expressed than the low-affinity cytochrome aa3 oxidase (Lamrabet et al. 2011). In our study, transcription of ctaD was actually up-regulated in both I. loihiensis (Fig. 1E) and M. daepoensis (Fig. 1F) upon transition to low O2 concentration, and remained at higher expression levels throughout the low O2 phase. Although transcript levels remained low, hardly reaching one transcript per gene (Figs S8C, S8D), the up-regulation was about 2- to 5-fold and significant (P < 0.05). A study of a Paracoccus denitrificans chemostat culture (Bosma et al. 1987) showed that the concentration of low-affinity terminal oxidase under fully oxic conditions was much larger than under O2-limited and anoxic conditions, which is in contrast to the up-regulation of low-affinity terminal oxidase gene expression under low O2-concentration conditions observed in this study. Overall, the regulation of terminal oxidase gene expression appears to be highly variable between organisms. The C-class HCOs have high affinity for O2, but their efficiency of energy conservation is lower than for the A-class HCOs. The stoichiometry of proton pumping is 1 H+/e− in A-class HCOs, whereas it is around 0.5 H+/e− in C-class HCOs (Han et al. 2011). Thus, the A-class HCOs require less O2, which is a limiting substrate under low O2-concentration conditions, than C-class HCOs, to generate the equivalent membrane potential (Han et al. 2011). Seen from the perspective of the individual bacterium, the competition is, however, not on resource efficiency but on fastest growth under any given condition, which in this case means the highest amount of proton pumping at any given O2 concentration. An optimized balance between electron flow through the low- and high-affinity terminal oxidases with given Km and maximum rate (Vmax) values will thus be determined by the diffusional supply of O2 to the population of terminal oxidases in the individual cell (Morris and Schmidt 2013). Electron flow through the low-affinity oxidases is preferential until O2 limitation due to kinetic parameters, amounts of terminal oxidases, and diffusion, results in a lowered electron flow that overall produces less proton export than by alternative flow with high-affinity terminal oxidases. The energetic advantage of A-class HCOs might explain the up-regulation of low-affinity terminal oxidase when the samples were exposed to low O2-concentration conditions. The low O2 conditions, however, also induced the up-regulation of expression of high-affinity terminal oxidase (cytochrome cbb3 terminal oxidase). Transcription of the putative low-affinity cytochrome bd-type oxidase, CIO The cytochrome bd-type oxidases are noteworthy for their high affinity for O2, especially the cytochrome bd-I terminal oxidase in E. coli, and resistance to inhibition by cyanide. They are essential for O2 consumption under micro-oxic conditions for bacteria such as E. coli (Cotter et al. 1990), D. vulgaris (Lamrabet et al. 2011), and Bacteroides fragilis (Baughn and Malamy, 2004). Further studies revealed a sub-group of quinol oxidases with a short Q-loop in subunit I and significantly higher resistance to cyanide than the initially classified classical cytochrome bd oxidases, named as cyanide-insensitive oxidases (CIO). CIOs are phylogenetically distinct from classical cytochrome bd oxidases (Miura et al. 2013). Based on recent oxygen affinity data and CIO phylogeny (Fig. S1), the CIO clades with Gluconobacter and Pseudomonas have low affinity for oxygen. Since the I. loihiensis CIO groups with Gluconobacter (Miura et al. 2013), and the Marinobacter CIO, although partially separated from Gluconobacter, is still most closely affiliated with this clade but far from other short Q-loop cytochrome bd oxidases (Fig. S1), we expected the CIOs in the two investigated strains to have low affinity for oxygen. The abundance of transcripts of these CIOs (cydA gene) was generally of the same order as the other low-affinity (cytochrome c) terminal oxidase in I. loihiensis (Figs S8C and S8E), while in M. daepoensis, transcripts of CIO were about 100 times more abundant than transcripts of cytochrome aa3 terminal oxidase (Figs S8D and S8F). This also means that, compared to I. loihiensis (Fig. S8E), the cydA gene was more highly expressed in M. daepoensis (Fig. S8F). The expression of the cydA gene in I. loihiensis (Fig. 1G) was significantly up-regulated (P < 0.05) after being exposed to low O2-concentration conditions (5 min, or even less). After prolonged low O2 concentration, the expression of cydA gene was down-regulated, and had the tendency to return to near-oxic levels, although the expression level was still significantly higher (P < 0.05) at the end of the incubation than under the initial oxic state. Interestingly, the expression of cydA was repressed significantly (P < 0.05) in M. daepoensis already at an O2 concentration of 5 µmol l−1 and after transition to low O2 conditions (Fig. 1H), with the exception of one of the three biological replicates (Fig. 1H: green dots); however, even in this case, cydA gene expression was down-regulated after 30 min under low O2 concentration. The repressed expression of CIO in M. daepoensis, and the fast down-regulation to the level of the fully oxic state after a brief up-regulation upon transition to low O2 conditions during the incubation of I. loihiensis, indicate that CIO expression may not be exclusively regulated by O2. As suggested for other bacterial species, CIO might have other crucial roles in physiology other than terminal oxidation. The characteristics of low affinity for O2 and low efficiency of energy conservation suggest that CIOs may act as a complementary enzyme when the HCOs are not functioning or are inhibited under stressful conditions (Arai et al. 2014). More generally, the up-regulation of all three types of terminal oxidases, regardless of their oxygen affinity, under nanomolar oxygen concentrations, could be a general stress response of the investigated strains because oxygen was the only electron acceptor in the system. The lessons learnt from the gene expression of terminal oxidases in pure cultures Although we can now describe the regulation of terminal oxidase transcription in response to nanomolar O2 concentrations, it is still not clear how the cells maintain a high respiratory rate when exposed to low O2 concentrations (Fig. S6): Do they replace the low-affinity with high-affinity terminal oxidases, do they increase the cellular concentration of high-affinity terminal oxidase without changing that of low-affinity terminal oxidases, or do they apply a combination of these two strategies? For example, in E. coli, both high-affinity terminal oxidases and low-affinity terminal oxidases are present in cells from oxygen-limited continuous cultures (Rice and Hempfling 1978). This lack of knowledge presents an additional limitation when interpreting environmental transcript data. Only one set of metatranscriptomic data from OMZs (off Chile and Peru) reported the distribution of genes and transcripts of terminal oxidases. The study showed that the abundances of transcripts for terminal oxidases were low, that the low-affinity terminal oxidase was expressed in all the sampling layers including the OMZ cores, and that the ratio of high-affinity terminal oxidases to low-affinity terminal oxidases was highest in samples from anoxic layers in terms of both genes and transcripts. However, the transcripts for low-affinity terminal oxidases were present in much higher numbers than those for high-affinity terminal oxidases (Kalvelage et al. 2015). The high number of transcripts for low-affinity terminal oxidases in OMZ samples may be due to the lack of genes for high-affinity terminal oxidases in many marine prokaryotes (Morris and Schmidt 2013), whereas the two strains investigated in our study have both types of genes and thus the ability to adapt to a low O2 environment by modulating their gene expression. This study shows that the pattern of expression of terminal oxidase genes is species-dependent, which illustrates the difficulty of connecting respiratory activities with gene expression, even in pure cultures, let alone in the environment. More detailed knowledge at combined transcript, proteome and activity levels is thus needed for meaningful interpretation of in situ transcriptome data in environmental samples (e.g. OMZs). SUPPLEMENTARY DATA Supplementary data are available at FEMSEC online. ACKNOWLEDGEMENTS We thank Susanne Nielsen, Preben G. Sørensen, Lars B. Pedersen, Trine Bech Søgaard, Britta Poulsen, and Anne Stentebjerg at Aarhus University for their technical assistance, Ian P. G. Marshall at Aarhus University for helpful discussions, and Minenosuke Matsutani and Kazunobu Matsushita at Yamaguchi University for the phylogenetic analysis. FUNDING This work was supported by the European Research Council [grant number 267233]. The funding agency had no role in study design, data collection and interpretation, or the decision to submit the work for publication. Conflicts of interest. None declared. REFERENCES Alvarez-Ortega C , Harwood CS . Responses of Pseudomonas aeruginosa to low oxygen indicate that growth in the cystic fibrosis lung is by aerobic respiration . Mol Microbiol . 2007 ; 65 : 153 – 65 . Google Scholar CrossRef Search ADS PubMed Arai H . Regulation and function of versatile aerobic and anaerobic respiratory metabolism in Pseudomonas aeruginosa . Front Microbiol . 2011 ; 2 : 103 . Google Scholar CrossRef Search ADS PubMed Arai H , Kawakami T , Osamura T et al. Enzymatic characterization and in vivo function of five terminal oxidases in Pseudomonas aeruginosa . J Bacteriol . 2014 ; 196 : 4206 – 15 . Google Scholar CrossRef Search ADS PubMed Baughn AD , Malamy MH . The strict anaerobe Bacteroides fragilis grows in and benefits from nanomolar concentrations of oxygen . Nature . 2004 ; 427 : 441 – 4 . Google Scholar CrossRef Search ADS PubMed Borisov SM , Lehner P , Klimant I . Novel optical trace oxygen sensors based on platinum(II) and palladium(II) complexes with 5,10,15,20-meso-tetrakis-(2,3,4,5,6-pentafluorphenyl)-porphyrin covalently immobilized on silica-gel particles . Anal Chim Acta . 2011 ; 690 : 108 – 15 . Google Scholar CrossRef Search ADS PubMed Borisov VB , Gennis RB , Hemp J et al. The cytochrome bd respiratory oxygen reductases . Biochim Biophys Acta - Bioenerg . 2011 ; 1807 : 1398 – 413 . Google Scholar CrossRef Search ADS Bosma G , Braster M , Stouthamer AH et al. Isolation and characterization of ubiquinol oxidase complexes from Paracoccus denitrificans cells cultured under various limiting growth conditions in the chemostat . Eur J Biochem . 1987 ; 165 : 657 – 63 . Google Scholar CrossRef Search ADS PubMed Brändén G , Gennis RB , Brzezinski P . Transmembrane proton translocation by cytochrome c oxidase . Biochim Biophys Acta . 2006 ; 1757 : 1052 – 63 . Google Scholar CrossRef Search ADS PubMed Buschmann S , Warkentin E , Xie H et al. The structure of cbb 3 cytochrome oxidase provides insights into proton pumping . Science . 2010 ; 329 : 327 – 30 . Google Scholar CrossRef Search ADS PubMed Calhoun MW , Thomas JW , Gennis RB . The cytochrome oxidase superfamily of redox-driven proton pumps . Trends Biochem Sci . 1994 ; 19 : 325 – 30 . Google Scholar CrossRef Search ADS PubMed Canfield DE , Kristensen E , Thamdrup B . Aquatic Geomicrobiology . San Diego : Elsevier Academic Press , 2005 . Comolli JC , Donohue TJ . Differences in two Pseudomonas aeruginosa cbb 3 cytochrome oxidases . Mol Microbiol . 2004 ; 51 : 1193 – 203 . Google Scholar CrossRef Search ADS PubMed Cotter PA , Chepuri V , Gennis RB et al. Cytochrome o (cyoABCDE) and d (cydAB) oxidase gene expression in Escherichia coli is regulated by oxygen, pH, and the fnr gene product . J Bacteriol . 1990 ; 172 : 6333 – 8 . Google Scholar CrossRef Search ADS PubMed D'mello R , Hill S , Poole RK . The cytochrome bd quinol oxidase in Escherichia coli has an extremely high oxygen affinity and two oxygen-binding haems: implications for regulation of activity in vivo by oxygen inhibition . Microbiology . 1996 ; 142 : 755 – 63 . Google Scholar CrossRef Search ADS PubMed Donachie SP . Idiomarina loihiensis sp. nov., a halophilic -Proteobacterium from the Lo'ihi submarine volcano, Hawai'i . Int J Syst Evol Microbiol . 2003 ; 53 : 1873 – 9 . Google Scholar CrossRef Search ADS PubMed Garcia-Horsman JA , Berry E , Shapleigh JP et al. A novel cytochrome c oxidase from Rhodobacter sphaeroides that lacks CuA . Biochemistry . 1994 ; 33 : 3113 – 9 . Google Scholar CrossRef Search ADS PubMed Garcia-Robledo E , Padilla CC , Aldunate M et al. Cryptic oxygen cycling in anoxic marine zones . Proc Natl Acad Sci USA . 2017 ; 114 : 8319 – 24 . Google Scholar CrossRef Search ADS PubMed Gong X , Garcia-Robledo E , Schramm A et al. Respiratory kinetics of marine bacteria exposed to decreasing oxygen concentrations . Appl Environ Microbiol . 2016 ; 82 : 1412 – 22 . Google Scholar CrossRef Search ADS Han H , Hemp J , Pace LA et al. Adaptation of aerobic respiration to low O2 environments . Proc Natl Acad Sci USA . 2011 ; 108 : 14109 – 14 . Google Scholar CrossRef Search ADS PubMed Helly JJ , Levin LA . Global distribution of naturally occurring marine hypoxia on continental margins . Deep Sea Res Part I Oceanogr Res Pap . 2004 ; 51 : 1159 – 68 . Google Scholar CrossRef Search ADS Hemp J , Robinson DE , Ganesan KB et al. Evolutionary migration of a post-translationally modified active-site residue in the proton-pumping heme-copper oxygen reductases . Biochemistry . 2006 ; 45 : 15405 – 10 . Google Scholar CrossRef Search ADS PubMed Jackson RJ , Elvers KT , Lee LJ et al. Oxygen reactivity of both respiratory oxidases in Campylobacter jejuni: the cydAB genes encode a cyanide-resistant, low-affinity oxidase that is not of the cytochrome bd type . J Bacteriol . 2007 ; 189 : 1604 – 15 . Google Scholar CrossRef Search ADS PubMed Kalvelage T , Lavik G , Jensen MM et al. Aerobic microbial respiration in oceanic oxygen minimum zones . PLoS One . 2015 ; 10 : e0133526 . Google Scholar CrossRef Search ADS PubMed Keefe RG , Maier RJ . Purification and characterization of an O2-utilizing cytochrome-c oxidase complex from Bradyrhizobium japonicum bacteroid membranes . Biochim Biophys Acta - Bioenerg . 1993 ; 1183 : 91 – 104 . Google Scholar CrossRef Search ADS Lamrabet O , Pieulle L , Aubert C et al. Oxygen reduction in the strict anaerobe Desulfovibrio vulgaris Hildenborough: characterization of two membrane-bound oxygen reductases . Microbiology . 2011 ; 157 : 2720 – 32 . Google Scholar CrossRef Search ADS PubMed Lehner P , Larndorfer C , Garcia-Robledo E et al. LUMOS - A sensitive and reliable optode system for measuring dissolved oxygen in the nanomolar range . PLoS One . 2015 ; 10 : e0128125 . Google Scholar CrossRef Search ADS PubMed Lever MA , Torti A , Eickenbusch P et al. A modular method for the extraction of DNA and RNA, and the separation of DNA pools from diverse environmental sample types . Front Microbiol . 2015 ; 6 : 476 . Google Scholar CrossRef Search ADS PubMed Miura H , Mogi T , Ano Y et al. Cyanide-insensitive Quinol Oxidase (CIO) from Gluconobacter oxydans is a unique terminal oxidase subfamily of cytochrome bd . J Biochem . 2013 ; 153 : 535 – 45 . Google Scholar CrossRef Search ADS PubMed Morris RL , Schmidt TM . Shallow breathing: bacterial life at low O2 . Nat Rev Microbiol . 2013 ; 11 : 205 – 12 . Google Scholar CrossRef Search ADS PubMed Pereira MM , Santana M , Teixeira M . A novel scenario for the evolution of haem–copper oxygen reductases . Biochim Biophys Acta - Bioenerg . 2001 ; 1505 : 185 – 208 . Google Scholar CrossRef Search ADS Pfaffl MW . A new mathematical model for relative quantification in real-time RT-PCR . Nucleic Acids Res . 2001 ; 29 : e45 . Google Scholar CrossRef Search ADS PubMed Pitcher RS , Watmough NJ . The bacterial cytochrome cbb 3 oxidases . Biochim Biophys Acta - Bioenerg . 2004 ; 1655 : 388 – 99 . Google Scholar CrossRef Search ADS Poole RK . Bacterial cytochrome oxidases: A structurally and functionally diverse group of electron-transfer protein . Biochim Biophys Acta - Rev Bioenerg . 1983 ; 726 : 205 – 43 . Google Scholar CrossRef Search ADS Pronk AF , Affourtit C , Mashingaidze C et al. The relationship between the in situ reduction level of the cytochrome c pool of Azorhizobium caulinodans growing in a chemostat with NH4+ or N2 as the N source and the total activity of cytochrome c oxidases . FEMS Microbiol Lett . 1995 ; 129 : 149 – 55 . Revsbech NP , Larsen LH , Gundersen J et al. Determination of ultra-low oxygen concentrations in oxygen minimum zones by the STOX sensor . Limnol Ocean Methods . 2009 ; 7 : 371 – 81 . Google Scholar CrossRef Search ADS Rice CW , Hempfling WP . Oxygen-limited continuous culture and respiratory energy conservation in Escherichia coli . J Bacteriol . 1978 ; 134 : 115 – 24 . Google Scholar PubMed Stam H , van Verseveld HW , de Vries W et al. Hydrogen oxidation and efficiency of nitrogen fixation in succinate-limited chemostat cultures of Rhizobium ORS 571 . Arch Microbiol . 1984 ; 139 : 53 – 60 . Google Scholar CrossRef Search ADS Stolper DA , Revsbech NP , Canfield DE . Aerobic growth at nanomolar oxygen concentrations . Proc Natl Acad Sci USA . 2010 ; 107 : 18755 – 60 . Google Scholar CrossRef Search ADS PubMed Sousa FL , Alves RJ , Ribeiro MA et al. The superfamily of heme–copper oxygen reductases: Types and evolutionary considerations . Biochim Biophys Acta - Bioenerg . 2012 ; 1817 : 629 – 37 . Google Scholar CrossRef Search ADS Wright JJ , Konwar KM , Hallam SJ . Microbial ecology of expanding oxygen minimum zones . Nat Rev Microbiol . 2012 ; 10 : 381 – 94 . Google Scholar CrossRef Search ADS PubMed Yoon J-H , Yeo S-H , Kim I-G et al. Marinobacter flavimaris sp. nov. and Marinobacter daepoensis sp. nov., slightly halophilic organisms isolated from sea water of the Yellow Sea in Korea . Int J Syst Evol Microbiol . 2004 ; 54 : 1799 – 803 . Google Scholar CrossRef Search ADS PubMed © FEMS 2018. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

FEMS Microbiology EcologyOxford University Press

Published: Apr 24, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off