Vol. 519: 47–59, 2015 MARINE ECOLOGY PROGRESS SERIES Published January 20 doi: 10.3354/meps11062 Mar Ecol Prog Ser Respiration quotient variability: bacterial evidence V. Romero-Kutzner1,*, T. T. Packard1, E. Berdalet2, S. O. Roy3, J.-P. Gagné4, M. Gómez1 1EOMAR, Grupo de Ecofisiología de Organismos Marinos, Universidad de Las Palmas de Gran Canaria 35017, Spain 2Institut de Ciències del Mar (CSIC), Passeig Marítim de la Barceloneta, 37-49, 08003 Barcelona, Spain 310 Ryan Court, Embrun, Ontario K0A 1W0, Canada 4Université du Québec à Rimouski, Rimouski, Québec G5L 3A1, Canada ABSTRACT: Respiratory metabolism was compared between 2 different physiological states of acetate- and pyruvate-grown cultures of Pseudomonas nautica and Vibrio natriegens. Here, we analyze 35 h and 520 h experiments in which time-courses of protein, pyruvate, acetate, respira- tory CO2 production (RCO2), respiratory O2 consumption (RO2), isocitrate dehydrogenase (IDH) activity, and potential respiration (Φ) were measured. Respiratory quotients (RQs) were calculated as the ratio of the respiration rates (RCO2/RO2). Such RQs are widely used in ocean ecosystem mod- els, in calculations of carbon flux, and in evaluations of the ocean’s metabolic balance. In all the cultures, the RQ tended to increase. In the case of P. nautica on acetate, the RQ rose nearly an order of magnitude from values below 1 during carbon-substrate sufficiency to values close to 10 during carbon-substrate deficiency. In all the cultures, the respiration rates during the growth period paralleled the biomass increase, but after the substrates were exhausted, the respiration rates fell. In contrast, through this same transition period, the IDH activity and the Φ remained rel- atively high for the first 10 h of carbon-substrate deprivation, and then, these enzyme activities fell slowly, along with the biomass, as the carbon-substrate deprivation continued. The nutritional state of the bacteria affected the RQ, rendering the RQ variable for physiological and ecological purposes. These results argue that ecosystem models, oceanographic calculations of carbon flux, and evaluations of the ocean’s metabolic balance that are influenced by bacterial metabolism need to be reconsidered in light of RQ variability. KEY WORDS: O2 consumption · CO2 production · Isocitrate dehydrogenase · IDH · Electron trans- port system · ETS · Potential respiration · Growth Resale or republication not permitted without written consent of the publisher INTRODUCTION investigator can differentiate lipid-based biological oxidation (metabolism) from carbohydrate-based me- The respiration quotient (RQ), the ratio of the CO2 tabolism. The majority of studies of respiration in produced to the O2 consumed in respiration, is an old aquatic ecosystems are based on RQ ranging from 0.7 concept, dating back to the 1860s. Armsby & Moul ton to 1.2 (Berggren et al. 2012), but in many ecological (1925) and Lusk (1928) discuss its use in animal hus- calculations of microbial respiration, an RQ of 1 is as- bandry and human physiology, presenting re search sumed (González et al. 2003, Bühring et al. 2006). of Pettenkofer & Voit (1866), Warburg (1926), and This practice occurs even though it is known that the others who were measuring RQ much earlier. RQ is a RQ varies from 0.65 to 1.4 in individual ecological dimensionless ratio, calculated mole per mole, and is microbial communities (del Giorgio et al. 2006). an index of the type of organic matter being oxidized Amado et al. (2013) argue that such an assumption is in respiration. From the shift in RQ from 0.7 to 1.0, an the only feasible option given the dearth of physio- *Corresponding author: [email protected] © Inter-Research 2015 · www.int-res.com 48 Mar Ecol Prog Ser 519: 47–59, 2015 logical studies that document the RQ variability in non-pathogenic marine bacterium discovered by bacterial respiration and they conclude that more in- Payne et al. (1961) in a Georgia salt-marsh (Lee & vestigations of RQ variability are needed. A recent Levy 1987, Lee 1995). Carbon-substrate limitation study of respiration in yeast found an RQ range of 0.4 conditions are common in the ocean as marine micro- to 1.4 (Slavov et al. 2014). When an organism is oxi- bial communities pass from bloom to post-bloom con- dizing carbohydrate, its CO2 production rate and its ditions (Liu et al. 2013), and are likely dominant in O2 consumption rate are equal, i.e. the RQ = 1. When oligotrophic ecosystems where bacterial respiration an organism is oxidizing protein, the RQ is around accounts for up to 59% of plankton respiration 0.8, and when oxidizing lipids, it is close to 0.7 (Can- (Robinson & Williams 2005). Here, we show that in tarow & Schepartz 1967, Guyton 1971, Hoar 1975, our cultures, the RQ can range higher than the upper Gnaiger 1983, Stanier & Forsling 1990). When burn- values reported above. To investigate this wide range, ing organic compounds richer in O2 than carbohy- we examine time-courses of the physiological respi- drates, such as carboxylic acids, and when converting ration rates (RO2 and RCO2) and the enzymatic activi- carbohydrate to fat, RQ values can be >1. When ties from the respiratory electron transport system metabolism consumes oxygen-rich oxalic acid (ETS) and isocitrate dehydrogenase (IDH). ETS and (C2H2O4), RQ can rise to 4 (Dilly 2001). At the other IDH activities are the biochemical enzyme activities end of the scale, gluconeogenesis (glucose synthesis) that largely control the RO2 consumption and RCO2 occurs when RQ falls below 0.7 (Cantarow & Schep- production (Packard et al. 1996a, Nelson & Cox 2005, artz 1967). In this way, RQ variations are generally Gnaiger 2009). Via aerobic respiration, marine or - explained by the fact that the amounts of CO2 and O2 ganisms obtain the energy to live from a wide range produced are dependent on the oxida tion state of the of compounds that are reduced in different, but well substrate and the pathways by which the substrate is coordinated biochemical pathways. Two of these key metabolized (Burton 1982, Kader 1987). Berggren et pathways are the Krebs or tricarboxylic acid (TCA) al. (2012) argue from their field studies of Quebec cycle and ETS. IDH is proposed to be a future ana- lakes that variability in RQ indicates shifts in bacterial lytical method for CO2 calculations in the ocean physiology and carbon consumption that cannot be (Packard et al. 1996a, Roy & Packard 2001), but it deduced from other measurements. However, in needs further investigation, improvement (Robinson oceanography, only a handful of RQ measurements & Williams 2005), and calibration. In our study, we have been made (Oviatt et al. 1986, Robinson et al. work with 2 substrates: acetate and pyruvate. Ace - 2002). Nevertheless, when the effort is made to deter- tate, a 2 carbon molecule, is transformed into acetyl- mine the RQ for each situation, as did Obernosterer et CoA at the entry of the TCA and, via IDH and alpha- al. (2008) in their Southern Ocean iron-fertilization keto glutarate dehydrogenase, loses both carbons as − experiment, an RCO2 production or a carbon oxidation CO2 and produces 8 reducing equivalents (e ) that in calculation is more accurate. turn, lead to the consumption of 2 molecules of O2 + − Whether in the laboratory or in the field, an RQ is (O2 + 4H + 4e → 2H2O) at cytochrome oxidase in essential in calculating either organic carbon con- the ETS. Accordingly, the potential RQ for acetate sumption or CO2 production from RO2 consumption would be 1.0. When pyruvate (a 3-carbon molecule) measurements (Boucher et al. 1994, Bergström 2011, is cycled through the TCA, it produces 10 reducing Giering et al. 2014). In this way, RQ becomes an equivalents and 3 molecules of CO2 and it consumes influential factor in ocean carbon-cycle studies, for 2.5 O2 molecules. This would result in an RQ of 1.2. calculating carbon flux from plankton metabolism Thus, the use of these different carbon substrates (Packard & Christensen 2004, Steinberg et al. 2008, becomes a tool to generate different values of RQ. Packard & Gómez 2013, Osma et al. 2014), and in In P. nautica, we found that acetate-grown cul - investigating whether the ocean is autotrophic or tures showed an RQ increase to 10 during carbon- heterotrophic (Ducklow & Doney 2013). Here, we limitation, whereas RQ in pyruvate-grown cultures measure the RQ in cultures of the marine bacteria remained near 1.0 (Roy et al. 1999). However, Pseudomonas nautica and Vibrio natriegens as they because of the continued lack of published physio- pass from carbon substrate sufficiency to carbon sub- logical information about RQ, we wanted to in - strate limitation. vestigate this topic with another species and over P. nautica is an oil-degrading bacterium from the longer periods of carbon-limitation. We expected that Gulf of Fos, France (Bonin et al. 1987a,b), and was different levels of carbon limitation and different used to advance studies of oil-spill bio-remediation durations of starvation would generate a wider range (Swannell et al. 1996). V. natriegens is a well-studied of RQs. Romero-Kutzner et al.: Respiration quotient variability in bacteria 49 MATERIALS AND METHODS triegens were experimentally established. They were grown on 400 mM NaCl, 10 mM MgSO4·7H2O, Experimental design 10 mM CaCl2·2H2O, 10 mM KCl, 25 mM NH4Cl, 0.33 mM phosphate buffer, 0.01 mM FeSO4·7H2O. To investigate the RQ in different bacterial growth Initial concentration of the culture-medium carbon- stages, time-course experiments were run on batch source was 30 mM sodium acetate or 20 mM pyru - cultures at 22°C, maintained on pyruvate or acetate vate. (Note that these concentrations would provide as described by Berdalet et al. (1995) and Packard et the same amount of organic carbon at the start of an al.