1 3

Arch Microbiol (2016) 198:149–159DOI 10.1007/s00203-015-1172-6

ORIGINAL PAPER

Dissolved inorganic carbon uptake in Thiomicrospira crunogena XCL‑2 is Δp‑ and ATP‑sensitive and enhances RubisCO‑mediated carbon fixation

Kristy J. Menning1 · Balaraj B. Menon2,3 · Gordon Fox1 · USF MCB4404L 2012 · Kathleen M. Scott1

Received: 20 August 2014 / Revised: 30 October 2015 / Accepted: 11 November 2015 / Published online: 18 November 2015 © Springer-Verlag Berlin Heidelberg 2015

were incubated in the presence of inhibitors targeting ATP synthesis (DCCD) or proton potential (CCCP). Incubations with each of these inhibitors resulted in diminished intra-cellular ATP, DIC, and fixed carbon, despite an absence of an inhibitory effect on proton potential in the DCCD-incu-bated cells. Electron transport complexes NADH dehydro-genase and the bc1 complex were found to be insensitive to DCCD, suggesting that ATP synthase was the primary target of DCCD. Given the correlation of DIC uptake to the intracellular ATP concentration, the ABC transporter genes were targeted by qRT-PCR, but were not upregulated under low-DIC conditions. As the T. crunogena genome does not include orthologs of any genes encoding known DIC uptake systems, these data suggest that a novel, yet to be identified, ATP- and proton potential-dependent DIC trans-porter is active in this bacterium. This transporter serves to facilitate growth by T. crunogena and other Thiomicrospi-ras in the many habitats where they are found.

Keywords Hydrothermal vent · Chemolithoautotroph · Carbon concentrating mechanism · Thiomicrospira · Calvin cycle · Carbon fixation

Introduction

Chemolithoautotrophs are primary producers at deep-sea hydrothermal vents, and their activities are particu-larly high where hydrothermal fluid mixes with seawater. Hydrothermal fluid is hot (>250 °C), acidic (pH 4), has elevated dissolved inorganic carbon (DIC) concentrations (~30 mM), and carries electron donors, including hydrogen sulfide (>4 mM; Childress et al. 1993). It is often diluted by seawater entrained in the crust before emission from the sea floor; dilute hydrothermal fluid temperature (2–40 °C)

Abstract The gammaproteobacterium Thiomicrospira crunogena XCL-2 is an aerobic sulfur-oxidizing hydro-thermal vent chemolithoautotroph that has a CO2 concen-trating mechanism (CCM), which generates intracellular dissolved inorganic carbon (DIC) concentrations much higher than extracellular, thereby providing substrate for carbon fixation at sufficient rate. This CCM presumably requires at least one active DIC transporter to generate the elevated intracellular concentrations of DIC measured in this organism. In this study, the half-saturation constant (KCO2) for purified carboxysomal RubisCO was measured (276 ± 18 µM) which was much greater than the KCO2 of whole cells (1.03 µM), highlighting the degree to which the CCM facilitates CO2 fixation under low CO2 conditions. To clarify the bioenergetics powering active DIC uptake, cells

Communicated by Erko Stackebrandt.

USF MCB4404L 2012 are: P. Wanjugi, M. Abdel-Rahim, M. B. Alak, L. J. Astatrjan, V. Bihary, F. Blazekovic, C. Cabrera, G. Camper, T. Chase, J. Dox, A. Echevarria, Q. A. Fisher, C. Georgeades, I. E. Heller, A. N. Hewlett, A. E. Justus, M. Kemp, M. Kondoff, J. P. Martin, E. McClenthan, G. R. Nicolas, J. Paoletti, S. Schuler, M. Skopis, S. R. Subar, E. R. Trebour.

Electronic supplementary material The online version of this article (doi:10.1007/s00203-015-1172-6) contains supplementary material, which is available to authorized users.

* Kathleen M. Scott kmscott@usf.edu

1 Department of Integrative Biology, University of South Florida, 4202 East Fowler Avenue, Tampa, FL 33620, USA

2 Department of Chemistry and Biochemistry, The University of Southern Mississippi, Hattiesburg, MS, USA

3 Present Address: Schepens Eye Research Institute, Harvard Medical School, Boston, MA, USA

150 Arch Microbiol (2016) 198:149–159

1 3

and chemical composition (pH ~ 6, 4 mM DIC, ~0.1 mM sulfide) reflect this mixing (Childress et al. 1993). As this dilute vent fluid flows into the cold (2–4 °C), alkaline (pH ~ 8), oxic seawater, turbulent eddies form, causing erratic conditions in which the concentration of the mol-ecules needed to support chemolithoautotrophy oscillate broadly over time. The concentration of CO2, for example, can fluctuate between 0.02 and 1 mM in a matter of sec-onds to days (Johnson et al. 1988; Goffredi et al. 1997). One adaptation to manage fluctuations in CO2 availability is a CO2 concentrating mechanism (CCM). CCMs allow autotrophic organisms to accumulate elevated intracellu-lar DIC concentrations when environmental availability of DIC is low (Dobrinski et al. 2005; Badger et al. 2006).

Thiomicrospira crunogena XCL-2 is an aerobic sulfur-oxidizing chemolithoautotrophic gammaproteobacterium with the first noncyanobacterial CCM to be characterized (Dobrinski et al. 2005). It has been isolated from hydro-thermal vents in both the Atlantic and Pacific Oceans and is one of the fastest growing chemolithoautotrophs known (Jannasch et al. 1985). This growth rate is high even when the concentration of DIC in the growth medium drops to 0.1 mM. Cells grown under these conditions can accumu-late intracellular DIC to 100× higher than extracellular in an energy-dependent manner, suggesting the presence of active DIC transporters (Dobrinski et al. 2005). The CCM in T. crunogena also includes a form IA RubisCO pack-aged into carboxysomes with a carboxysomal carbonic anhydrase encoded by csoSCA (Scott et al. 2006; Dobrin-ski et al. 2010, 2012). CsoSCA appears to facilitate carbon fixation but not DIC uptake (Dobrinski et al. 2010) and is found in an operon with the genes encoding the carboxy-some shell proteins and form IA RubisCO (Scott et al. 2006); form II RubisCO, as well as a second, presum-ably noncarboxysomal form IA RubisCO, are encoded elsewhere on the chromosome and are preferentially tran-scribed under high CO2 conditions (Dobrinski et al. 2012).

Carboxysomes are a key component of the CCMs in cyanobacteria; however, it is the arsenal of DIC-transport-ers that are responsible for generating the elevated intra-cellular DIC concentrations that facilitate carboxysome function (Badger et al. 2006; Price et al. 2009). In cyano-bacteria, there are three families of well-characterized DIC transporters currently known to contribute to CCMs. Under severe DIC limitation, the primary, ATP-hydrolyzing ABC transporter BCT1, encoded by cmpABCD, is expressed in freshwater cyanobacteria (Badger et al. 2006; Price et al. 2009). The Na+-dependent secondary transporter, BicA, is a member of the SulP family, which has medium-to-low affinity for bicarbonate (Price et al. 2004). BicA is induc-ible in Synechococcus sp. PCC 7002, but appears to be con-stitutively expressed in Synechocystis sp. PCC 6803 (Price et al. 2004; Eisenhut et al. 2007). Another high-affinity,

inducible Na+–HCO3− symporter, SbtA, is expressed by

cyanobacteria under low-DIC conditions (Espie and Kan-dasamy 1994; Shibata et al. 2002).

In addition to these bicarbonate transporters, many cyanobacteria have CO2 hydration complexes associ-ated with the thylakoid or cell membrane. Two forms of this complex have been described: NDH-I3 (induc-ible, high affinity) and NDH-I4 (low affinity, constitutively expressed) (Shibata et al. 2001; Maeda and Price 2002). Both are believed to couple the hydration of CO2 to exer-gonic electron transfer to the quinone pool; coupling the hydration to an exergonic redox reaction precludes the pos-sibility of the complex catalyzing the dehydration back-reaction (Price 2011). The electron donor powering these complexes is unknown. The function of these complexes is believed to convert any noncarboxysomal, cytoplasmic CO2 to HCO3

− to prevent it from diffusing out of the cell; presumably, the rate of carboxysomal uptake and fixation of the CO2 ‘recaptured’ by the CO2 hydration complexes exceeds the rate of abiotic HCO3

− dehydration back to CO2 (Price 2011). These complexes consist of many of the pro-teins of the respiratory NAD(P)H dehydrogenase complex NDH-I (NdhA–C, NdhE, NdhG–O; (Battchikova and Aro 2007). Cyanobacteria whose genomes encode CO2 hydra-tion complexes have several paralogous copies of genes encoding NdhD and NdhF proteins (e.g., ndhD1, ndhD2, ndhD3, ndhD4). The NdhD3 and NdhF3 proteins are part of the NDH-I3 complex, while NdhD4 and NdhF4 occupy the NDH-I4 complex (Battchikova and Aro 2007). CupA (NDH-I3 complex) and CupB proteins (NDH-I4 complex) are unique to CO2 hydration complexes and may function in CO2 hydration (Price 2011).

The T. crunogena genome encodes many transporters, including 10 apparent operons for ABC influx transport-ers composed of genes for an ATP-hydrolyzing subunit, transmembrane subunit, and a solute-binding subunit. The remaining ATP-dependent transporters appear to be involved in protein secretion, H+, Na+, or other cation efflux (Scott et al. 2006). Additionally, fifty-eight sec-ondary transporters representing 28 transporter families, including one SulP family transporter, are present in the T. crunogena genome. Oligonucleotide microarrays with probes designed for all of the genes in the T. crunogena genome did not identify any transporters whose gene tran-scripts were more abundant when cells were cultivated under low-DIC conditions (Dobrinski et al. 2012).

T. crunogena constructs carboxysomes and generates elevated intracellular DIC concentrations when grown under low-DIC conditions (Dobrinski et al. 2005, 2012); the objective of this study was to elucidate the degree to which transmembrane DIC uptake facilitates carbon fixa-tion and illuminate the nature of the contribution to DIC uptake by cellular bioenergetics (ATP, as well as proton

151Arch Microbiol (2016) 198:149–159

1 3

potential, (Δp), comprised of membrane potential (ΔΨ) and ΔpH).

Materials and methods

Carboxysomal RubisCO was purified and its half-satura-tion constant (KCO2) and Vmax were measured, in order to compare the enzyme half-saturation constant to that previ-ously measured in whole cells. To determine whether DIC uptake is powered by proton potential or ATP hydrolysis, T. crunogena cells were incubated with a protonophore (CCCP) or an ATP synthase inhibitor (DCCD). To clarify whether the correlation between intracellular ATP concen-trations and DIC uptake was due to the activity of an ABC transporter, prior microarray results for ABC transporter genes (Dobrinski et al. 2012) were verified via qRT-PCR to determine whether their transcript levels were sensitive to the abundance of DIC during growth, as might be expected for a DIC transporter.

Reagents

Tritiated water, NaH14CO3, 14C-methylamine hydro-

chloride, and 14C-D-sorbitol were purchased from MP Biomedicals. The metabolic inhibitors carbonyl cya-nide m-chlorophenylhydrazone (CCCP) and N,N′-dicyclohexylcarbodiimide (DCCD) were obtained from Sigma. 14C-tetraphenylphosphonium bromide was pur-chased from American Radiolabeled Chemicals, Inc.

Cultivation of T. crunogena

T. crunogena XCL-2 (Jannasch et al. 1985) was culti-vated at room temperature (~20 °C) in 40 mM thiosulfate-supplemented artificial seawater (TASW, pH 8) as previ-ously described (Dobrinski et al. 2005). For carboxysomal RubisCO purification, an 8 L batch culture was grown to late exponential phase before harvesting. For investigating the effects of metabolic inhibitors on DIC uptake and fixa-tion, T. crunogena was cultivated under DIC limitation in TASW in chemostats (~0.1 mM DIC in the growth cham-ber; New Brunswick Scientific BioFlo 110) at a dilution rate of 0.1 h−1. A dO2/pH controller was used to control aeration with O2 to maintain culture oxygen concentrations (~20–50 μM), and titration with 10 N KOH to maintain the pH (=8). To determine whether transcript abundance for solute-binding components of ABC transporters was sensitive to the DIC concentration present during growth, cells were cultivated both in DIC-limited chemostats (as described above) and high-DIC chemostats (50 mM DIC, dilution rate of 0.1 h−1, sparged with 5 % v/v CO2 in O2 as needed; (Dobrinski et al. 2005).

Carboxysomal RubisCO purification

T. crunogena cells (8 L) were harvested and resuspended in 15 ml of TEMB buffer (10 mM TRIS, 1 mM EDTA, 10 mM MgSO4, 20 mM NaHCO3

−) containing 2 mg ml−1 lysozyme, and 0.2 μM phenylmethylsulfonyl fluoride. The cell suspension was inverted several times after addition of 15 ml of nonionic detergent B-PER II (Pierce Scien-tific). The cell slurry was then sonicated (Branson model 450 sonifier, constant duty cycle, power output = 7, 10 s). A 150 μl portion of 1 mg ml−1 bovine pancreatic DNase I solution in TEMB buffer was added. The sonicated cell lysate was then agitated for 30 min at room temperature (~20 °C, Reliable Scientific D55 tilting platform shaker, 60 cycles per min) and centrifuged at 10,000×g for 10 min. The resulting cell pellet was resuspended in 30 ml TEMB buffer, which was then sonicated (4 × 30 s bursts, 1 min cooling intervals between bursts). Nonidet P-40 (final con-centration = 1 % v/v) was added to the sonicated lysate to solubilizate membranes and stirred for 1 h at room tem-perature. The sonicated lysate was centrifuged at 10,000×g for 10 min. The resulting supernatant was centrifuged at 48,000×g for 30 min, and the resulting pellet was resus-pended in 3 ml TEMB buffer and centrifuged at 1000×g for 3 min. The supernatant from this centrifugation was highly enriched in carboxysomes and was loaded onto a 36 ml 10–60 % w/v continuous sucrose gradient in TEMB buffer prepared using a gradient former (Bethesda Research Lab-oratories). Gradients were centrifuged in a swinging bucket rotor (Beckman JS 28.38, 104,000×g, 30 min, 4 °C). A milky white band of purified carboxysomes formed at the middle of the gradient and was harvested. After adjusting the volume to ~36 ml with TEMB buffer, the carboxysome suspension was centrifuged at 150,000×g for 2 h at 4 °C in a Type 70Ti rotor (Beckman). The resulting pellet was resuspended in 1 ml TEMB buffer and stored at 4 °C until further use.

To purify RubisCO from these carboxysome prepara-tions, they were freeze-thawed multiple times to disrupt the carboxysome shell and release the RubisCO. Ruptured car-boxysome shells were removed by centrifugation (~30 min, 4 °C, 14,000×g). RubisCO purity was verified via SDS-PAGE; gels were stained with coomassie brilliant blue to visualize proteins (Sambrook and Russell 2001).

RubisCO activity was determined via 14CO2 fixation as in Scott et al. (2007). To verify whether carboxylase activity was proportional to the amount of enzyme present, activity was assayed at a range of protein concentrations (0–13 μg). To determine the pH range favorable for RubisCO activ-ity, it was necessary to remove the influence of pH on the CO2 concentration (e.g., RubisCO activity would appear elevated at low pH and diminished at high pH due to ele-vated CO2 abundance at lower pH). To accomplish this, an

152 Arch Microbiol (2016) 198:149–159

1 3

isotopic disequilibrium method was used, as described in (Scott et al. 2004).

To determine the KCO2 and Vmax, an assay buffer pH of 7.5 was chosen to accommodate higher CO2 concentrations (it was anticipated that the KCO2 value would be high), while still maintaining a pH value favorable to enzyme activity. Four independent RubisCO assays were performed using the same RubisCO preparation. The concentrations of CO2 present in these incubations were calculated from the pH values and DIC concentrations as in Yokota and Kitaoka (1985). KCO2 and Vmax were calculated from the carbon fixation rates and CO2 concentrations using least-squares nonlinear regression as implemented in XLStatistics 10.05.30 (http://www.deakin.edu.au/~rodneyc/XLStatistics; XLent Works, Australia).

Effect of metabolic inhibitors on DIC uptake and fixation

Cells were cultivated in chemostats under DIC limitation (TASW growth medium; 0.1 mM DIC, 7.6 mM (NH4)2SO4 (Dobrinski et al. 2005). One culture was used to conduct pilot experiments to determine appropriate concentrations and exposure times for the inhibitors. Three independent cultures were used to measure the impact of the inhibitors on DIC uptake.

Cultures were harvested via centrifugation (5000×g, 4 °C, 10 min). Thiosulfate was removed by washing the cell pellet three times with thiosulfate-free and DIC-free TASW medium (pH 8). The cell pellet was resuspended in thiosulfate-free and DIC-free TASW (OD600 ~ 4) and bubbled with CO2-free air for ~15–30 min until the DIC concentration was undetectable by gas chromatography (Dobrinski et al. 2005).

This cell suspension was divided into three aliquots: one was incubated with 1 mM of the ATP synthase inhibitor DCCD, the second aliquot was incubated with 10 μM of the protonophore CCCP, and the third portion was without an inhibitor, but was amended with DMSO to 0.1 % v/v to act as a solvent control, as both the DCCD and CCCP solu-tions were prepared in this solvent. Cell suspensions were incubated on ice for 1 h under CO2-free air before use in incubations. These inhibitor concentrations and exposure times were determined by pilot experiments to reliably result in an effect on ATP, ΔΨ, and intracellular pH (data not shown).

Intracellular pH, ΔΨ, intracellular ATP, and DIC uptake and fixation were measured for cells incubated at room temperature (~20 °C) under four conditions: (1) in the absence thiosulfate (2) presence of thiosulfate (40 mM), (3) 10 µM CCCP (plus thiosulfate) or (4) 1 mM DCCD (plus thiosulfate). All incubations had a final concentration of 0.1 % v/v DMSO to control for the solvent added with the DCCD or CCCP.

To determine whether the inhibitors were affecting their targets (CCCP: intracellular pH and ΔΨ; DCCD: [ATP]), intracellular pH, ΔΨ, and cellular ATP concentrations were measured. Silicone oil centrifugation (SOC) was used to measure intracellular pH and ΔΨ; for this technique, cells incubated with appropriate radiolabel were pipetted into screw-cap 0.6-ml centrifuge tubes which had been pre-loaded with 20 μl of a dense killing solution (1:2 v/v triton: 1 M glycine, pH 3 for incubations with 14C-methylamine, pH 10 for all others) overlain with 65 μl silicone oil (Dow Chemicals SF1156) (Dobrinski et al. 2005). When this three-layered system is centrifuged (40 s, maximum speed), the cells carry their contents through the silicone and into the killing solution. The tube is immediately frozen with liquid nitrogen, and the bottom layer (containing the cells) is clipped into a scintillation tube with scintillation cocktail to quantify the radiolabel contained in the cells.

To measure intracellular pH via SOC, cells were incu-bated in TASW supplemented with DMSO (solvent control) or inhibitors dissolved in DMSO (10 μM CCCP or 1 mM DCCD), and 2 mM (specific activity 56 mCi mmol−1) 14C-methylamine as described in Dobrinski et al. (2005). ΔΨ was measured via SOC, using tetraphenylphospho-nium bromide (TPP), which diffuses across the cell mem-brane and accumulates in the cytoplasm in proportion to the ΔΨ (Rottenberg 1979; Olsson et al. 2003). Incuba-tion conditions were identical to those noted above, but were amended with 14C-tetraphenylphosphonium bro-mide to a concentration of 20 mM with a specific activity of 5 mCi mmol−1. ATP was measured in cells incubated as above for SOC. A 20 µl portion of each cell suspension was added to 400 µl of incubation buffer. After 30 s, 10 μl of the suspension was stirred into 90 μl of 95 °C distilled water and incubated at 95 °C for 2 min in a thermocycler, after which it was cooled on ice and stored at −80 °C. ATP was quantified four times for each treatment using a com-mercially available bioluminescence kit (SIGMA #FL-AA) and luminometer (Promega, Glomax 20/20).

DCCD has been shown to inhibit electron transport in the NDH-1 and cytochrome bc1 complexes in some organ-isms (Yagi 1987; Wang et al. 1998), so it was necessary to verify that the primary target for this inhibitor in T. cruno-gena cells was the intracellular ATP pool and not cel-lular reductant levels (e.g., NAD/NADH or ubiquinone/ubiquinol). Cell membrane vesicles were prepared from T. crunogena cultivated under DIC limitation, using a proto-col modified from (Sinegina et al. 2005). Approximately 100 mg of cells were suspended in 5 ml of buffer contain-ing 50 mM HEPES at pH 7, 100 mM KCl, 0.5 mM EDTA, and 0.5 mM PMSF. Cells were sonicated until lysed and centrifuged at 10,000 g for 20 min at 4 °C. The supernatant was centrifuged at 35,000 g for 30 min, the resulting pel-leted cell membrane vesicles were resuspended in 500 µl

153Arch Microbiol (2016) 198:149–159

1 3

of buffer containing 25 mM bis-tris-propane at pH 6 and 10 mM betaine, and were flash-frozen using liquid nitrogen and stored at −80 °C.

NDH-1 activity was assayed, three times for each treat-ment (Sinegina et al. 2005), with the following modi-fied procedure. Thawed membrane vesicles (20 µl) were added to 1 ml of buffer containing 25 mM HEPES (pH 7.5), 200 µM NADH, and 3.5 mM NaCN, and substrate (50 µM decylubiquinone). To test for inhibition by DCCD, 1 mM DCCD was added to the sample and incubated on ice for 1 h. NADH oxidation was followed at 340 nm (ε = 6.2 mM−1 cm−1). To measure any NADH oxida-tion not mediated by NDH-1 (other enzymes, or nonbio-logical reactions), parallel assays were conducted without ubiquinone.

Cytochrome bc1 complex activity was assayed simi-larly to Rotsaert et (al. 2008) by the reduction of 50 µM cytochrome c in a buffer containing 50 mM KH2PO4 (pH 6), 250 mM sucrose, 0.2 mM EDTA, 0.1 % BSA, and 50 µM decylubiquinol. Decylubiquinol was synthe-sized by dissolving decylubiquinone in ethanol and add-ing to a buffer (50 mM KH2PO4 pH 6, 250 mM sucrose, 0.2 mM EDTA, 1 mM NaCN, and 0.1 % BSA) to a con-centration of 620 µM (Trumpower and Edwards 1979). Sodium borohydride was added to a concentration of 5.3 mM and incubated for 1 h until H2 bubbles stopped forming. Cell membrane vesicles were incubated for 1 h with and without 1 mM DCCD. Cytochrome bc1 complex activity was monitored by tracking cytochrome c reduc-tion (ε = 21.5 mM−1 cm−1 at 550 nm) in the presence of 50 µM decylubiquinol. As for the NDH-1 complex, cytochrome c reduction was verified to be due to the bc1 complex, and not alternative enzymatic or chemical reac-tions, by repeating the assay in the absence of decylubiqui-nol. Cytochrome c reduction was measured three times for each treatment (without substrate, uninhibited with sub-strate, and with substrate and inhibitor). Both NDH-1 and cytochrome bc1 activities were measured at room tempera-ture (~20 °C).

Solute‑binding protein transcript abundance

All genes from the T. crunogena genome likely to encode solute-binding proteins for ABC transporters were col-lected from TransportDB (http://www.membranetransport.org; (Ren et al. 2007). Solute-binding protein genes that were present in apparent operons with other genes encod-ing the other components of ABC transporters (transmem-brane and ATP-binding subunits) were selected, and their transcript abundance under low- versus high-DIC growth conditions was quantified (see below). In two cases, two genes encoding solute-binding proteins were present in the apparent operons (Tcr_0981 and Tcr_0982; Tcr_1152 and

Tcr_1153). Both genes in both operons were assayed for transcript abundance.

One solute-binding protein gene (Tcr_1153) had sequence similarity to those involved in cyanobacterial HCO3

− as well as NO3−-uptake (Fig. S1). Typically, T.

crunogena is cultivated with ammonium as a nitrogen source (see above). To elucidate whether this gene (and others) was responsive to the concentration of DIC or to NO3

−, cells were cultivated in chemostats under three dif-ferent conditions: (1) DIC-limited (DIC = 0.1 mM, 13 mM NaNO3), (2) NO3

−-limited (DIC = 50 mM, 0.5 mM NO3

−), (3) NH3-limited (DIC = 50 mM; NH3 = 0.5 mM). Once cultures had reached steady state, they were harvested by centrifugation (5 min, 10,000×g, 4 °C), flash-frozen in liquid nitrogen, and stored at −80 °C until use.

RNA was isolated from these cells using the RiboPure-Bacteria kit (Ambion, Inc.), and relative transcript abun-dance was determined via SYBR Green RT-PCR, using the QuantiTect SYBR Green RT-PCR kit (Qiagen, Inc.), and primers listed in Table 1. The ΔΔCT method was used to compare the relative amounts of transcripts in RNA iso-lated from cells grown in different chemostats, using 16S transcript levels as the calibrator (Livak and Schmittgen 2001; Dobrinski et al. 2012). The csoSCA gene, which encodes a carboxysomal carbonic anhydrase, has higher transcript abundances in DIC-limited cells (Dobrinski et al. 2010, 2012). Therefore, transcript abundance for this gene was determined in parallel with the solute-binding pro-tein genes, as a positive control for CCM induction under DIC-limited growth. When gene transcript abundance was compared for cells cultivated in the presence of different nitrogen sources (NH3 vs. NO3

−), the gene encoding a sol-ute-binding protein from an ABC transporter (Tcr_2079) was used as a negative control, as it is part of a gene cluster encoding the enzymes for metabolizing phosphonate (Scott et al. 2006), making it unlikely to be sensitive to nitrogen sources.

Statistical analysis

For the effects of inhibitors on pH, ΔΨ, ATP, NDH-1, and cytochrome bc1 complex, treatment groups were compared using a one-way ANOVA in SPSS Statistics 19, with a sig-nificance level of α ≤ 0.05. To analyze the effects of inhibi-tors on DIC uptake and fixation, the results from the three independent cultures were combined by treating them as an experiment with randomized block design, with subsam-ples within treatments. Each culture was a block, and there were four treatments per block (−thiosulfate, +thiosulfate, CCCP, DCCD). Since the data were too heteroscedastic for analysis of variance to be appropriate, generalized least squares, implemented in R, was used on natural log-trans-formed data (Amemiya 1985; Walker and Smith 2009).

154 Arch Microbiol (2016) 198:149–159

1 3

Table 1 Fold change in transcript abundance for genes encoding ABC transporter solute-binding proteins in Thiomicrospira crunogena

Locus taga Predicted substrate(s) qRT-PCR primers (prod-uct location on gene)b

Comparisonc Nutrient in commond Fold change in transcript abundance (95 % CI)

Tcr_0033 NO3−, RSO3

−, HCO3− TGCCGATGAAAGC-

CAAAGTGGTTCTTGGTTGCTGACGGG-

TAAGGGTAA(911–1052)

Low/high DIC NO3− 1.01 (0.52–1.95)

Tcr_0544 Phosphate AGCAGCTGTT-GTCTCTAACCCTGT

TGCCCATAGCGTCAT-CAAGTTTGC

(36–174)

Low/high DIC NO3− 1.10 (0.06–2.03)

Tcr_0577 Zn+2 GCGTTGGTTCAGC-CGTATTTGGAT

ACGCTTTGAGATGA-GACGGACGAA

(88–193)

Low/high DIC NO3− 0.94 (0.56–1.59)

Tcr_0981 ? AGCTTCCTTTGGA-TACGTCAGCCT

AATACACATCGTCTG-CACCGGGTT

(278–370)

Low/high DIC NO3− 1.12 (0.061–2.05)

Tcr_0982 Toluene tolerance GACGAAAGCACAG-CAAGATGCGTT

ATTGCCATCT-GCTTGGGTGACTTC

(267–438)

Low/high DIC NO3− 0.94 (0.37–2.37)

Tcr_1152 NO3− ATAGTGGTTGTGC-

CACCCTCAAGTAGTGACAGCTGTGC-

CGCATTTAAC(493–609)

Low/high DIC NO3− 1.12 (0.07–1.77)

Tcr_1153 NO3− AAGCCGGAACGATC-

GAAGGGTATTTGCGGATCGTCGT-

GTTGGGATATT(686–859)

Low/high DIC NO3− 0.86 (0.65–1.14)

Tcr_1153 NO3− (Primers listed above) NO3

−/NH3 High DIC 73 (41–132)

Tcr_1182 Dipeptides AATTTCTTCGCGC-CGATTCCTTGG

AAACCGGATTCGG-GATAGAAGCCA

(715–908)

Low/high DIC NO3− 0.66 (0.30–1.49)

Tcr_1800 NO3−, RSO3

−, HCO3 TACAGTGATTGGC-CAGGATGGGTT

ATCGATTTGTC-CAGCTGCGAATGC

(94–231)

Low/high DIC NO3− 0.54 (0.39–0.74)

Tcr_1927 MoO4−2 TTGCACATGCCTTC-

CAATACGTGGAAACCGGAC-

TATTCGGGT-CAACCA

(500–589)

Low/high DIC NO3− 0.90 (0.73–1.12)

Tcr_2079 Phosphonate GCCATGATTATTGC-CGCAGACGTT

ATAGGGCAGGAT-CAATAGCGCCTT

(472–646)

Low/high DIC NO3− 0.59 (0.19–1.81)

155Arch Microbiol (2016) 198:149–159

1 3

For qRT-PCR, Ct values from three aliquots of an RNA sample were determined, and the average and standard deviation were calculated. Error was propagated from the Ct values using standard statistical methods.

Results

Carboxysomal RubisCO purification and Michaelis–Menten kinetics

Carboxysomes (Fig. 1a) and carboxysomal RubisCO (Fig. 1b) were successfully purified from T. crunogena. Two bands are apparent after SDS-PAGE of enzyme released from carboxysomes by freeze-thaw treatment (Fig. 1b); their molecular weights are consistent with those predicted from the sequences of the genes encoding the large (51.9 kDa; cbbL) and small (13.4 kDa; cbbS) subunits of T. crunogena carboxysomal RubisCO.

Assay carbon fixation rates were dependent on the amount of purified RubisCO added (Fig. S1) and were zero when ribulose 1,5-bisphosphate was omitted. RubisCO activity was highest from pH 7–8.5 (Fig. 1c). The KCO2 of the purified carboxysomal RubisCO was 276 µM (SD ± 18) and the Vmax was 252 nmol CO2 min−1 mg pro-tein−1 (SD ± 7; Fig. 1d). Based on the SDS-PAGE gel and sequences of cbbL and cbbS, the molecular mass of the

holoenzyme (L8S8) is 5.22 × 105 g mol−1. From this, and factoring in the presence of eight active sites per holoen-zyme, the kcat for this enzyme is 0.27 s−1.

Effect of metabolic inhibitors on DIC uptake and fixation

The expected impacts of CCCP and DCCD on their target parameters in intact cells (intracellular pH, ΔΨ, and ATP) were apparent (Table 2). Thiosulfate presence alkalinized intracellular pH, brought ΔΨ to more negative values, and increased ATP, as expected given that T. crunogena can use thiosulfate as a donor for its electron transport chain. Rela-tive to thiosulfate-energized cells, the addition of the proto-nophore CCCP diminished intracellular pH and drove ΔΨ to more positive values, as expected, due to proton entry into the cells. ATP concentrations were low in the presence of this inhibitor, possibly due to diminishment of the proton potential necessary to drive ATP synthase. DCCD addition also diminished intracellular ATP concentrations, likely due to direct interaction of this inhibitor with ATP synthase.

DCCD did not inhibit the activities of NADH dehydro-genase or the cytochrome bc1 complex (Table 3), suggest-ing that its major impact on the cells was due to its inter-action with ATP synthase and not secondary effects due to inhibition of the activities of electron transport chain complexes.

Table 1 continued

Locus taga Predicted substrate(s) qRT-PCR primers (prod-uct location on gene)b

Comparisonc Nutrient in commond Fold change in transcript abundance (95 % CI)

Tcr_2079 Phosphonate (primers listed above) NO3−/NH3 High DIC 0.45 (0.36–0.57)

Tcr_2149 Metal ions ACGACCCTCAGAAT-GAAGCGAGTT

TGCAATCCTCGCAC-CTCCATATCA

(443–629)

Low/high DIC NO3− 1.18 (0.79–1.77)

Tcr_0841 (csoSCA) N/A TCTAAGGCAGACC-CTACACATCAA

CGCCGCTTTATGGT-CATCACT

(759–819)

Low/high DIC NO3− 232 (165–327)

16S N/A CGAATATGCTCTACG-GAGTAAAGGT

CGCGGGCTCATCCTT-TAG

(109–163)

(qRT-PCR calibrator)

a IMG gene locus tag designationsb Primers are presented F first, and R secondc Cells were grown in chemostats. Low DIC chemostats were DIC-limited (0.1 mM) and had high nitrogen concentrations (13 mM NO3

− or NH3); high DIC (50 mM) chemostats were nitrogen-limited (0.5 mM NO3

− or NH3)d ‘Nutrient in common’ indicates that the two populations of cells that were compared with respect to transcript abundance either used the same nitrogen source (‘NO3

−’) or were both cultivated under high-DIC conditions (‘high DIC’)

156 Arch Microbiol (2016) 198:149–159

1 3

Inorganic carbon uptake and fixation by T. crunogena cells is stimulated by the presence of thiosulfate (Fig. 2, Fig. S2) as previously noted (Dobrinski et al. 2005). When thiosulfate was added, cells generated intracellular

DIC concentrations ~15× higher than extracellular; in its absence, it was only 2×. Both inhibitors reduced the amounts of fixed carbon, which is consistent with the Cal-vin-Benson cycle’s requirements for ATP, diminished in the

Fig. 1 Thiomicrospira crunogena carboxysomal RubisCO purifi-cation and characterization. a Transmission electron micrograph of purified carboxysomes. Samples were stained with 1 % w/v ammo-nium molybdate and observed under a transmission electron micro-scope at ×30,000 magnification. b SDS-PAGE analysis of RubisCO purified from carboxysomes. Left lane molecular weight standards (Fisher BioReagents EZ-Run Prestained Rec Protein Ladder). Right lane 14 μg purified carboxysomal RubisCO, stained with coomas-sie brilliant blue. c Response of RubisCO activity to assay pH. Error

bars indicate the standard error of the CO2 fixation rates. d Response of RubisCO activity to the concentration of CO2. Results from four independent experiments are shown with different symbols. The curve is the rectangular hyperbola resulting from incorporating the aver-age KCO2 (276 μM) and Vmax (252 nmol min−1 mg protein−1) from the four experiments into the Michaelis–Menten equation. For d, rate measurement precision was very high; consequently, error bars indicating the SE of the CO2 fixation rates are very short and are obscured by the symbols

Table 2 Physiological parameters for DIC-limiteda Thiomicrospira crunogena cells treated with inhibitorsb

a Extracellular DIC = 0.1 mMb Mean values are provided ±SD (n = 4)c Values with different superscripts (a, b, c) are significantly different from each other (Scheffe test, α = 0.05)

Treatment 40 mM Na2S2O3 Intracellular pHc ΔΨ (mV) ATP (mM)

No inhibitor − 7.15 ± 0.04a −94 ± 6a 0.36 ± 0.04a

No inhibitor + 7.67 ± 0.05b −131 ± 2b 4.46 ± 0.92b

10 μM CCCP + 7.13 ± 0.04a −105 ± 2c 0.32 ± 0.08a

1 mM DCCD + 7.77 ± 0.22b −127 ± 2b 1.45 ± 0.11c

157Arch Microbiol (2016) 198:149–159

1 3

presence of either CCCP or DCCD (Table 2), and NAD(P)H, which is presumably diminished by CCCP due to inhibi-tion of reverse electron transport caused by collapse of the cellular proton potential. Both inhibitors also diminished the concentrations of intracellular DIC; cells treated with CCCP had intracellular DIC concentrations ~1.5× higher than extracellular, while those incubated with DCCD had a 3.5× gradient (Fig. 2).

ABC transporter solute‑binding protein transcript abundance

None of the genes encoding solute-binding proteins, which were present near genes encoding the membrane and ATP-binding components of ABC transporters, demonstrated a measurable increase in transcript abundance when cells were grown under conditions inducing the CCM (Table 1). It was anticipated that if any were involved in DIC uptake, their transcript levels would be higher under low-DIC con-ditions. The positive control, encoding the carboxysomal carbonic anhydrase (csoSCA), demonstrated a >200-fold increase in transcript abundance under DIC-limiting growth

conditions. Transcripts of Tcr_1153, which is homologous both to the cognate genes of the cyanobacterial bicarbonate and nitrate ABC transporters, were not responsive to DIC concentrations during growth (Table 1). Instead, Tcr_1153 transcripts were more abundant when nitrate was present as the nitrogen source than with ammonium. As expected, the gene from a likely phosphonate transporter was insensitive to both DIC abundance and nitrogen source (Table 1).

Discussion

A comparison of enzyme and whole cell KCO2 values is consistent with active DIC transport by T. crunogena. The carboxysomal RubisCO KCO2 (276 µM) was much higher than the KCO2 of whole cells (1.03 µM; (Dobrinski et al. 2005). Indeed, it is rather high for a form I RubisCO, which typically ranges from 22 to 180 μM in bacteria (Viale et al. 1990; Larimer and Soper 1993; Hernandez et al. 1996; Horken and Tabita 1999; Badger and Bek 2008) but is on par with other form IA and IB RubisCOs found in organ-isms with carboxysomes and CCMs (173–293 μM, with one extremely high value of 750 μM by Prochlorococ-cus marinus RubisCO; (Badger 1980; Badger et al. 1998; Scott et al. 2007; Dou et al. 2008; Marcus et al. 2011). In the absence of active uptake, the KCO2 for whole cells should be similar to or greater than that of the carboxyso-mal RubisCO since the intracellular CO2 will be similar to or lower than extracellular. The kcat (0.27 s−1) is lower than that measured for other enzymes (0.4–11.4 s−1; (Badger 1980; Badger et al. 1998; Scott et al. 2007; Dou et al. 2008; Marcus et al. 2011), which may indicate that some of the proteins had been inactivated during the complex enzyme purification process (e.g., multiple centrifugations, and car-boxysome freeze-thaw).

DIC accumulation by T. crunogena cells is sensitive both to the proton gradient and ΔΨ, as well as to intracel-lular ATP concentrations (Table 2; Fig. 2). Collapsing the proton gradient and ΔΨ with CCCP had the most dramatic effect on DIC uptake and fixation; the addition of DCCD also had a marked effect on both parameters. Given that CCCP collapses the proton potential and also diminishes the intracellular ATP concentration, the effect of this inhibi-tor on DIC uptake and fixation could be due to either of

Table 3 Thiomicrospira crunogena NDH-1 and cytochrome bc1 complex activity in the absence and presence of DCCD

a Substrates are 50 μM decylubiquinone or decylubiquinol for NDH-1 and cytochrome bc1 complexes, respectively

NDH-1 (µmol s−1 ±SD) Cytochrome bc1 (µmol s−1 ±SD)

−Substratea 31.2 (n = 1) 0.1 ± 0.4 (n = 3)

+Substrate 272 ± 48 (n = 3) 3.6 ± 0.3 (n = 3)

+Substrate +1 mM DCCD 309 ± 41 (n = 3) 3.4 ± 0.3 (n = 3)

Fig. 2 Effects of thiosulfate and metabolic inhibitors on the concen-trations of intracellular fixed carbon and dissolved inorganic carbon (DIC) in Thiomicrospira crunogena cells incubated in the presence of 0.1 mM DIC. ‘−TS’ = 0 mM thiosulfate, ‘+TS’ = 40 mM thio-sulfate, ‘+CCCP’ = 40 mM thiosulfate + 10 μM carbonyl cyanide m-chlorophenyl hydrazine, ‘+DCCD’ = 40 mM thiosulfate + 1 mM N,N′-dicyclohexylcarbodiimine. Error bars indicate 95 % confidence intervals; lowercase letters over the error bars indicate significant differences (Tukey’s test, p < 0.05) among fixed carbon (a, b, c) and intracellular DIC (d, e) values

158 Arch Microbiol (2016) 198:149–159

1 3

these effects. Given that DCCD specifically targets ATP synthase, it appears that intracellular DIC accumulation is sensitive to intracellular ATP concentrations when the pro-ton potential and intracellular pH are still intact.

Multiple models for coupling proton potential and/or ATP to DIC uptake are possible (Fig. S3). The mechanism(s) for DIC uptake by T. crunogena are likely to depart from those described for cyanobacteria, as genes orthologous to those encoding cyanobacterial DIC uptake systems are not present in the T. crunogena genome (Table S1). Given the apparent sensitivity of DIC uptake to ATP concentrations, an ABC transporter seemed a likely candi-date. In the T. crunogena genome, the genes that were the best match for those encoding BCT1, the cyanobacterial ABC transporter responsible for bicarbonate uptake, prob-ably encode a nitrate transporter as they were responsive to NO3

− availability in the growth medium (Table 1) and are in an apparent operon with genes encoding enzymes nec-essary for assimilatory nitrate reduction (Fig. S4). Nitrate uptake by this system is further supported by phylogenetic analysis of the T. crunogena gene encoding the solute-bind-ing component of this putative ABC transporter; the gene falls within a well-supported clade with other solute-bind-ing components whose neighboring genes encode nitrate reductase, and distinct from the clade containing the cyano-bacterial solute-binding proteins from the BCT1 bicarbo-nate transporter (Fig. S4).

Reliance on an ATP-sensitive transporter could be assumed to be more energetically expensive in T. cruno-gena than utilizing a secondary transporter. Genome data are consistent with T. crunogena relying on the periplasmic Sox system to oxidize thiosulfate to sulfuric acid (Scott et al. 2006), resulting in periplasmic proton accumula-tion (Friedrich et al. 2001). The majority of the electrons abstracted from thiosulfate by the Sox system are trans-ferred to the terminal oxidase (cytochrome cbb3 oxidase; (Scott et al. 2006); the cbb3 complex contributes to pro-ton potential by using these electrons to consume cyto-plasmic protons as it reduces O2 to H2O, while also acting as a vectoral proton pump (Rauhamaki et al. 2012). This two-component electron transport chain (Sox complex and cytochrome cbb3 oxidase) is likely responsible for generat-ing cellular proton potential during lithotrophic growth by T. crunogena with thiosulfate as the electron donor (Scott et al. 2006). Intuitively, the most efficient strategy for DIC uptake would be to use a secondary transporter to directly couple this Δp to uptake. Instead, it appears that the energy stored in the transmembrane proton potential is communi-cated to the DIC uptake system by using the Δp to power ATP synthase. In general, energy conversion results in some energy loss as heat. However, given that ATP syn-thase appears to convert virtually all of the energy from proton entry into the cell to ADP phosphorylation (Toyabe

et al. 2011), energy losses incurred with this added step may be minimal.

A comparison of the carboxysomal RubisCO KCO2 to the whole cell KCO2 makes it clear that DIC transport is essential to the function of the CCM in T. crunogena but the identity of the transporter or transporters responsible for HCO3

− uptake is unclear. The presence of carbox-ysomes, which are also found in cyanobacteria suggests that chemolithoautotrophs like T. crunogena have CCMs that function similarly to those that have been character-ized in cyanobacteria. Once identified, the genes encoding the DIC uptake system(s) in noncyanobacterial autotrophs are certain to provide insight into the ecophysiology and evolution of DIC uptake in autotrophic microorgan-isms across the many phyla and habitats where they are present.

Acknowledgments We are tremendously thankful to the National Science Foundation for their support of this project (NSF-MCB-0643713 to K.M.S.). We would also like to thank Gordon Can-non and Sabine Heinhorst for helpful discussions and for the use of their facilities for carboxysome studies, and the anonymous review-ers for their insightful suggestions, which improved the quality of this manuscript.

References

Amemiya T (1985) Advanced econometrics. Harvard University Press, Cambridge

Badger MR (1980) Kinetic properties of ribulose 1,5-bisphosphate carboxylase/oxygenase from Anabaena variabilis. Arch Bio-chem Biophys 201:247–254

Badger MR, Bek EJ (2008) Multiple RubisCO forms in proteobacte-ria: their functional significance in relation to CO2 acquisition by the CBB cycle. J Exp Bot 59:1525–1541

Badger MR et al (1998) The diversity and coevolution of RubisCO, plastids, pyrenoids, and chloroplast-based CO2-concentrating mechanisms in algae. Can J Bot 76:1052–1071

Badger MR, Price GD, Long BM, Woodger FJ (2006) The environ-mental plasticity and ecological genomics of the cyanobacterial CO2 concentrating mechanism. J Exp Bot 57:249–265

Battchikova N, Aro E (2007) Cyanobacterial NDH-1 complexes: multiplicity in function and subunit composition. Physiol Plant 131:22–32

Childress JJ et al (1993) Inorganic carbon uptake in hydrothermal vent tubeworms facilitated by high environmental pCO2. Nature 362:147–169

Dobrinski KP, Longo DL, Scott KM (2005) A hydrothermal vent chemolithoautotroph with a carbon concentrating mechanism. J Bacteriol 187:5761–5766

Dobrinski KP, Boller AJ, Scott KM (2010) Expression and function of four carbonic anhydrase homologs in the deep-sea chemo-lithoautotroph Thiomicrospira crunogena. Appl Environ Microb 76:3561–3567

Dobrinski KP, Enkemann SA, Yoder SJ, Haller E, Scott KM (2012) Transcription response of the sulfur chemolithoautotroph Thi-omicrospira crunogena to dissolved inorganic carbon limitation. J Bacteriol 194:2074–2081

Dou Z, Heinhorst S, Williams E, Murin E, Shively J, Cannon G (2008) CO2 fixation kinetics of Halothiobacillus neapolitanus mutant

159Arch Microbiol (2016) 198:149–159

1 3

carboxysomes lacking carbonic anhydrase suggest the shell acts as a diffusional barrier for CO2. J Biol Chem 283:10377–10384

Eisenhut M, Schubert EVWLJH, Bauwe BWIH (2007) Long-term response toward inorganic carbon limitation in wild type and glycolate turnover mutants of the cyanobacterium Synechocystis sp. strain PCC 6803. Plant Physiol 144:1946–1959

Espie GS, Kandasamy RA (1994) Monensin inhibition of Na+-dependent HCO3− transport distinguishes it from Na+-inde-pendent HCO3− transport and provides evidence for Na+/HCO3− symport in the cyanobacterium Synechococcus UTEX 625. Plant Physiol 104:1419–1428

Friedrich CG, Rother D, Bardischewsky F, Quentmeier A, Fischer J (2001) Oxidation of reduced inorganic sulfur compounds by bac-teria: emergence of a common mechanism? Appl Environ Micro-biol 67:2873–2882

Goffredi SK, Childress JJ, Desaulniers NT, Lee RW, Lallier FH, Ham-mond D (1997) Inorganic carbon acquisition by the hydrother-mal vent tubeworm Riftia pachyptila depends upon high external P-CO2 and upon proton-equivalent ion transport by the worm. J Exp Biol 200:883–896

Hernandez JM, Baker SH, Lorbach SC, Shively JM, Tabita FR (1996) Deduced amino acid sequence, functional expression, and unique enzymatic properties of the form I and form II ribulose bisphosphate carboxylase/oxygenase from the chemoautotrophic bacterium Thiobacillus denitrificans. J Bacteriol 178:347–356

Horken K, Tabita FR (1999) Closely related form I ribulose bis-phosphate carboxylase/oxygenase molecules that possess dif-ferent CO2/O2 substrate specificities. Arch Biochem Biophys 361:183–194

Jannasch H, Wirsen C, Nelson D, Robertson L (1985) Thiomicrospira crunogena sp. nov., a colorless, sulfur-oxidizing bacterium from a deep-sea hydrothermal vent. Int J Syst Bacteriol 35:422–424

Johnson KS, Childress JJ, Hessler RR, Sakamoto-Arnold CM, Bee-hler CL (1988) Chemical and biological interactions in the Rose Garden hydrothermal vent field. Deep Sea Res 35:1723–1744

Larimer FW, Soper TS (1993) Overproduction of Anabaena 7120 rib-ulose-bisphosphate carboxylase/oxygenase in Escherichia coli. Gene 126:85–92

Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2-ΔΔCT method. Methods 25:402–408

Maeda S, Price MRBGD (2002) Novel gene products associated with NdhD3/D4-containing NDH-1 complexes are involved in photo-synthetic CO2 hydration in the cyanobacterium, Synechococcus sp PCC7942. Mol Microbiol 43:425–435

Marcus Y, Altman-Gueta H, Wolff Y, Gurevitz M (2011) Rubisco mutagenesis provides new insight into limitations on photo-synthesis and growth in Synechocystis PCC6803. J Exp Bot 62:4173–4182

Olsson K, Keis S, Morgan H, Dimroth P, Cook G (2003) Bioenergetic properties of the thermoalkaliphilic Bacillus sp. strain TA2.A1. J Bacteriol 185:461–465

Price GD (2011) Inorganic carbon transporters of the cyanobacterial CO2 concentrating mechanism. Photosynth Res 109:47–57

Price G, Woodger F, Badger M, Howitt S, Tucker L (2004) Identifica-tion of a SulP-type bicarbonate transporter in marine cyanobac-teria. Proc Natl Acad Sci USA 101:18228–18233

Price GD, Badger MR, Woodger FJ, Long BM (2009) Advances in understanding the cyanobacterial CO2-concentrating-mecha-nism (CCM): functional components, Ci transporters, diversity, genetic regulation and prospects for engineering into plants. J Exp Bot 59:1441–1461

Rauhamaki V, Bloch DA, Wikstrom M (2012) Mechanistic stoichi-ometry of proton translocation by cytochrome cbb(3). Proc Natl Acad Sci USA 109:7286–7291. doi:10.1073/pnas.1202151109

Ren Q, Chen K, Paulsen IT (2007) TransportDB: a comprehensive database resource for cytoplasmic membrane transport systems and outer membrane channels. Nucl Acids Res 35:D274–D279

Rotsaert F, Ding M, Trumpower B (2008) Differential efficacy of inhibition of mitochondrial and bacterial cytochrome bc1 com-plexes by center N inhibitors antimycin, ilicicolinH and funicu-losin. Biochim Biophys Acta 1777:211–219

Rottenberg H (1979) The measurement of membrane potential and ΔpH in cells, organelles, and vesicles. In: Colowick SP, Kaplan NO (eds) Methods in enzymology. Elsevier, New York, pp 547–569

Sambrook J, Russell DW (2001) Molecular cloning: A laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor

Scott KM, Schwedock J, Schrag DP, Cavanaugh CM (2004) Influence of form IA RubisCO and environmental dissolved inorganic car-bon on the δ13C of the clam-bacterial chemoautotrophic symbio-sis Solemya velum. Environ Microbiol 6:1210–1219

Scott KM et al (2006) The genome of deep-sea vent chemolithoauto-troph Thiomicrospira crunogena. PLoS Biol 4:1–17

Scott KM, Henn-Sax M, Longo D, Cavanaugh CM (2007) Kinetic isotope effect and biochemical characterization of form IA RubisCO from the marine cyanobacterium Prochlorococcus marinus MIT9313. Limnol Oceanogr 52:2199–2204

Shibata M et al (2001) Distinct constitutive and low-CO2-induced CO2 uptake systems in cyanobacteria: genes involved and their phylogenetic relationship with homologous genes in other organ-isms. Proc Natl Acad Sci 98:11789–11794

Shibata M et al (2002) Genes essential to sodium-dependent bicarbo-nate transport in cyanobacteria. J Biol Chem 277:18658–18664

Sinegina L, Wikstrom M, Verkhovsky M, Verkhovskaya M (2005) Activation of isolated NADH:ubiquinone reductase I (com-plex I) from Escherichia coli by detergent and phospholipids. Recovery of ubiquinone reductase activity and changes in EPR signals of iron-sulfur clusters. Biochemistry 44:8500–8506

Toyabe S, Watanabe-Nakayama T, Okamoto T, Kudo S, Muneyuki E (2011) Thermodynamic efficiency and mechanochemical coupling of F-1-ATPase. Proc Natl Acad Sci USA 108:17951–17956. doi:10.1073/pnas.1106787108

Trumpower B, Edwards C (1979) Purification of a reconstitutively active iron-sulfur protein (oxidation factor) from succinate cytochrome c reductase complex of bovine heart mitochondria. J Biol Chem 254:8697–8706

Viale AM, Kobayashi H, Akazawa T (1990) Distinct properties of Escherichia coli products of plant-type ribulose-1,5-bisphos-phate carboxylase/oxygenase directed by two sets of genes from the photosynthetic bacterium Chromatium vinosum. J Biol Chem 265:18386–18392

Walker AFZENINJ, Smith AASGM (2009) Mixed effects models and extensions in R. Springer, New York

Wang Y, Obungu V, Beattie D (1998) Dicyclohexylcarbodiimide inhibits proton pumping in ubiquinol:cytochrome c oxidoreduc-tase of Rhodobacter sphaeroides and binds to aspartate-187 of cytochrome b1. Arch Biochem Biophys 352:193–198

Yagi T (1987) Inhibition of NADH-ubiquinone reductase activity by N, N’-dicyclohexylcarbodiimide and correlation of this inhi-bition with the occurrence of energy-coupling site 1 in various organisms. Biochemistry 26:2822–2828

Yokota A, Kitaoka S (1985) Correct pK values for dissociation con-stant of carbonic acid lower the reported Km values of ribulose bisphosphate carboxylase to half. Presentation of a nomograph and an equation for determining the pK values. Biochem Bio-phys Res Commun 131:1075–1079