Hydrobiologia 523: 1–7, 2004. 2004 Kluwer Academic Publishers. Printed in the Netherlands.
A simple and highly reproducible technique to extract the 14CO2 resulting from respiration of 14C-labeled seawater samples Christian Tamburini* & Marc Tedetti Laboratoire de Microbiologie, Geochimie et Ecologie Marines (LMGEM) – UMR 6117 CNRS/INSU – Universite´ de la Me´diterrane´e, Centre d’Oce´anologie de Marseille, Campus de Luminy – Case 901, 163 Avenue de Luminy, F-13288 Marseille cedex 09, France (*Author for correspondence: Tel.: +33-4-91829049, Fax: +33-4-91829051, E-mail: [email protected]
) Received 24 May 2002; in revised form 11 June 2003; accepted 28 January 2004
Key words: Bacteria, respiration, mineralization,
CO2 extraction, carbon dioxide, seawater
Abstract Production of 14CO2 from water samples amended with 14C-labeled molecules allows estimation of the mineralization rates of a variety of organic compounds. Diverse protocols have been already proposed for 14 CO2 extraction and trapping. Yet, eﬃciency of carbon dioxide recovery greatly varies and, despite the application of operators, data from duplicate samples are often divergent. In order to propose a standardized protocol and to avoid the main source of artifacts, we suggest keeping the trapping agent for 14 CO2 directly into a 20 ml scintillation vial hanging in the degassing bottle. To validate this protocol, we plotted the radioactivity (disintegration per minute) due to 14CO2 recovered after acidiﬁcation of 42 seawater samples supplemended with sodium (14C)-bicarbonate (NaH14CO3) against the actual radioactivity of the corresponding added bicarbonate solutions. Eﬃciency of this protocol results satisfactory; as for these 42 assays the percentage of recovery is equal to 99% with a very low variability (±1%, p ¼ 0.05).
Introduction The increased demand for carbon dioxide production in the Ocean has stressed the need for measurements of bacterial respiration rates. Different methodologies were proposed to provide such data as exhaustively reviewed by Williams (1984). The most frequently used methods ﬁt in two categories. A ﬁrst one relies on the possibility to follow the decrease of O2 concentrations in water samples during deﬁned incubation periods (Langdon, 1993). Presently available techniques are suﬃciently precise to operate on natural samples of surface seawater. It is an uncontested advantage to obtain respiration rates unbiased by any enrichment of the sample. Conversely, such experiments report only the overall production of carbon dioxide. To diﬀerentiate the part actually caused by bacterial respiration, previous ﬁltration
of the sample (elimination of organisms other than bacteria), or addition of speciﬁc agents inhibiting the respiration of organisms other than bacteria are necessary. Moreover, sensitivity of presently available methods does not allow to measure respiration rates in intermediate and deep-sea water layers. A second category of methods relies on the supplement of samples with radiolabeled organic compounds. Respiration rates of the added substrates are commonly estimated by measuring the carbon dioxide produced during deﬁned incubation periods of treated seawater samples. These techniques suﬀer from the necessity to know the actual concentration of the organic substrate in the studied samples to be able to calculate the actual metabolic rate, the reason why most of the data
2 are expressed as potential respiration rates of the added substrate (Williams, 1984). However, these techniques may be helpful to study the potential microbial respiration rates through the whole oceanic system, including deep-sea water and sediments, on a large diversity of organic substrates. Labile organic compounds (i.e. expected to be used as carbon and energy sources by a large diversity of bacterial species) like 14C-labeled glucose, glutamic acid, leucine and amino acid mixture are the most frequently used substrates (Williams, 1970; Dawson & Gocke, 1978; Bianchi et al., 1998, 1999; Unanue et al., 1998, 1999; Tamburini et al., 2003). More complex compounds like radiolabeled membranes and soluble proteins (Nagata et al., 1998), or like 14C-chitin (Kirchman & White, 1999) are also used to evaluate the fate of less labile material. The ability to produce 14CO2 from diﬀerent 14 C-labeled hydrocarbons is currently performed to measure the microbial capacity to degrade petroleum compounds (Seki, 1976; Walker & Colwell, 1976; Foght et al., 1989; Piehler et al., 1999; Rooney-Varga et al., 1999). Several techniques have been used to trap carbon dioxide from the culture. These methods suﬀer from poor reproducibility, respiration rates measured on duplicate subsamples being frequently diverging. Furthermore, these methods cannot be considered as standardized, as actual eﬃciency of 14 CO2 recovery is frequently not determined. The aim of this study was to design a simple standardized protocol to provide reproducible 14 C-compound respiration rates, that is easy to operate on large set of samples on board research vessel and that requires only usual disposable glassware.
Materials and methods Current methodology used to cultivate seawater samples
The current methodology used to cultivate 14Clabeled seawater samples is brieﬂy called to mind hereafter. A solution of the selected 14C-labeled substrate was aseptically added to a deﬁned volume of seawater, usually comprised between 20
and 200 ml, depending on the expected microbial activity. Final substrate concentration was used either at trace concentration to calculate the actual respiration rate when the natural concentration is known, either at saturation level to measure the maximum velocity (Wright & Hobbie, 1965). So, ﬁnal substrate concentration is commonly included in the range 1–1000 nM. The seawater samples were incubated in a tightly closed sterile bottle in the dark at in situ temperature, and when possible under in situ pressure conditions (Tamburini et al., 2003). Incubation period should be within the linear phase of utilization of the added compound, that is commonly between 20 min and 12 h. During the incubation period, carbon dioxide resulting from respiration processes remains trapped as HCO 3 , due to the buﬀering capacity of seawater. At the end of the incubation period, acidiﬁcation of the samples induced a 14CO2 release in the bottle headspace. Then, 14CO2 was recovered with a trapping agent, followed by measurements of the radioactivity by scintillation counting. Practitioners of these measurements know that, whatever their application, recovery of produced CO2 suﬀers from poor reproducibility, with respiration rates from duplicate subsamples frequently diverging. Furthermore, these methods cannot be considered as standardized, as actual eﬃciency of 14CO2 recovery is frequently not determined. Several techniques have been used to trap carbon dioxide from the culture. Most of techniques used a piece of accordion-folded chromatography paper to increase the trapping agent surface. This piece of paper is in a small plastic cup (Hobbie & Crawford, 1969), or in an open scintillation vial hanging from the stopper of the culture bottle (Dawson & Gocke, 1978). Other authors suggest using a glass (or plastic) cup to contain the trapping agent (Kuparinen & UusiRauva, 1980). Alternative techniques resort to a bubbling circuit to extract the 14CO2, and then to trap it in a barium hydroxide solution (Williams & Askew, 1968) or in phenethylamine (Garabe´tian, 1991). Speciﬁc protocol suggested to enhance, simplify and standardize the procedure of 14CO2 recovery Several series of experiments using seawater samples previously ﬁxed by adding 1% (ﬁnal concen-
3 tration) of buﬀered formaldehyde was performed. Use of ﬁxed seawater during this methodological approach was to avoid biological variability between assays. For each series, ﬁxed seawater samples were distributed in polycarbonate bottles (Narrow-mouth square bottle, Nalgene) holding a Teﬂon covered stir bar (7 · 36 mm). The volume of the used bottle varied depending on the sample volume (see below). These bottles were screwcapped with septum closure (Thermoplastic elastomere septum, Nalgene). Each of the samples was supplemented with NaH14CO3 solution (CEA Corp., 55 mCi mmol)1 speciﬁc activity). As an improvement of the 14CO2 trapping method, we suggest to keep the trapping agent directly into a 20 ml scintillation vial, hanging up in the culture bottle. The scintillation vial is suspended from the cap of the culture bottle by the way of a ﬁsh-hook at the end of a nylon yarn, as shown in Figure 1. Ethanolamine was chosen as CO2 absorbent, its trapping property being better than that of phenethylamine (Kuparinen & Uusi-Rauva, 1980). The Nalgene bottles were tightly closed, and then seawater was acidiﬁed by injecting 1 ml HCl 6 N through the cap, using a 1 ml syringe ﬁtted with a thin needle. At these conditions (pH ~1.5), the aqueous H-14CO)3 evolved to gaseous 14CO2. The bottles were set on a 15-position electromagnetic stirrer, allowing gentle agitation (300 rpm) at room temperature. At the end of the stirring period, the scintillation vial was taken from the
sample bottle and accurately wiped with a paper handkerchief moisten with a decontaminating solution (RBS 25, Chemical Products). Ten milliliter of Ready Safe (Packard) scintillation cocktail were added, and the radioactivity was counted by using a Packard 1600 TR liquid scintillation counter. Data were corrected against those from control samples. Equal volume of NaH14CO3 solution used for experiments was directly placed into four scintillation vials and counted together with the samples to provide a value of actual radioactivity injected in all experimental bottles. Checking out for the optimum incubation period The appropriate incubation time for eﬀective extraction of 14CO2 was checked within a period between 10 and 12 h. An intermediate concentration of NaH14CO3 corresponding to 3500 disintegrations per min (DPM) was used. Checking out for diﬀerent volumes of samples An additional set of samples ranging in volume between 20 and 130 ml, commonly used for various environmental respiration assays (from the sea surface down to the deep waters in an oligotrophic area) was analyzed. Checking out for adequacy between sample volume and headspace volume of the extraction bottle volume Two diﬀerent assays including 20 ml samples in 125 ml bottles and 40 ml samples in 250 ml bottles were performed along with the general experiments done with 70 ml samples in 500 ml bottles. Strategy used to check for the eﬃciency and reproducibility of the suggested protocol
Figure 1. Schematic diagram of a reaction bottle used to extract carbon dioxide from seawater sample.
To prove the eﬃciency and reproducibility of the proposed protocol, 10 series of experiments differed in amount of NaH14CO3 added (1000–8000 DPM) were performed. The chosen range of radioactivity corresponds to the respiration values usually obtained for water column in oligotrophic areas. For each series, six 70 ml samples were distributed in 250 ml Nalgene bottles.
4 Results Extraction period allowing optimal
Figure 2 presents the eﬃciency of carbon dioxide retrieval from 70 ml seawater samples during increasing degassing periods (15, 30 min, 1, 2, 4, 8 and 12 h) using 250 ml bottles (Nalgene). Eﬃciency is clearly time-dependent, and stable recovery near 100% occurs after 4 h. Inﬂuence of sample and extraction bottle volumes on CO2 recovery
extracted for 4 h in 250 ml vials) recovering the range of DPM usually obtained during 14C-compound respiration assays from the sea surface to the deep waters. Results indicate that eﬃciency of carbon dioxide extraction was 99% with a determination coeﬃcient of 0.99. 14CO2 recovery is also highly reproducible with a variation of ±1% (for a variation coeﬃcient of 95%; n ¼ 42).
Table 1 shows that, using the proposed protocol, eﬃciency of 14CO2 recovery remains adequate, within a range of sample volumes between 20 and 130 ml, so long as the degassing process is in the same extraction bottle. Conversely, changing the bottle’s volume modiﬁes the eﬃciency of the protocol. Reducing the headspace in the bottle and increasing the incubation period, result in an improved 14CO2 recovery (over 90%). As a consequence, the actual extraction eﬃciency should be determined for each kind of extraction bottle. Alternatively, when using extraction bottles of the same volume, it should be possible to determine and use a unique percentage of recovery for the whole series of samples. Eﬃciency and reproducibility of
Figure 3 reports the eﬃciency of 14CO2 extraction from a set of 42 assays (70 ml of seawater sample
The present protocol is a modiﬁcation of techniques previously described to recover the 14CO2 released by acidiﬁcation of 14C-labeled culture Table 1. The eﬀect of sample volume and extraction bottle volume on the eﬃciency of 14CO2 extraction process. Extraction was at room temperature with a stirring speed of 300 rpm. Extraction bottles were square polycarbonate Nalgene bottles Sample
Incubation Percent recovery
volume (ml) bottle volume (ml) time (h)
90 ± 1 (n = 6)
93 ± 2 (n = 6)
77 ± 2 (n = 6)
77 ± 1 (n = 62)
82 ± 1 (n = 6)
99 ± 1 (n = 42) 80 ± 2 (n = 6)
95 ± 1 (n = 6)
78 ± 1 (n = 6)
97 ± 1 (n = 6)
Figure 2. Eﬃciency of carbon dioxide recovery during extraction periods ranging between 10 min and 12 h. Seawater samples (70 ml in 250 ml bottles) were labeled with a NaH14CO3 solution (4100 DPM per sample), extraction was at room temperature with a stirring speed of 300 rpm.
5 (Dawson & Gocke, 1978). The suggested modiﬁcation is to keep the trapping agent (i.e. 1 ml ethanolamine) directly into the scintillation vial, hanging up in the culture bottle, while in the current protocol ethanolamine is adsorbed on a small strip of chromatography paper kept in a handling cup or directly suspended in the atmosphere of the degassing vessel (Dawson & Gocke, 1978; Deming, 1993). Use of the referred technique frequently introduces several artefacts. Some are due to the variability of the actual area of each strip of paper for ethanolamine absorption, and as follows the 14 CO2 trapping eﬃciency. Another source of variability is the contact of the paper strip with cup or the bottle wall, leading to a loss of 14CO2-loaded ethanolamine, and so an underestimation of respiration rates. Moreover, the paper strip might be contaminated by 14C-labeled seawater droplets generated by movements of the magnetic stirrer. Due to the speciﬁc activity of the tracer, contamination of the paper strip by even small droplets of labeled seawater is suﬃcient to cause a large overestimate of the DPM count. Another degassing protocol is to ﬂush the 14 CO2 by bubbling unlabeled CO2 (or N2) through the sample, and then through two serial trapping vials containing a trapping cocktail. This protocol appears eﬃcient (Garabe´tian, 1991), but variability between two replicates may be important. Gas leaks may originate underestimation of the respiration rates. Conversely, contamination of the blowing circuit by labeled seawater aerosols produced by accidental over bubbling may induce drastic overestimation of the measured rates.
The proposed protocol avoids all these sources of artefacts: 1. Reproducibility of the trapping agent volume is deﬁned by the precision of the used volumetric micropipette. The absorbing surface is also highly reproducible, variability of the diameter of the used scintillation vials being lower than measurable. 2. Contamination of the trapping agent by droplets (or aerosols) of 14C-labeled seawater is unlikely, since the trapping agent is protected by the scintillation vial. Even if the outer wall of the scintillation vial was contaminated by labeled seawater droplets, such contamination should be eliminated by the wiping of the vials before their introduction in the scintillation counter. Moreover, being outside of the scintillation cocktail, such a contamination should escape to scintillation counting. 3. Suppression of the ﬂushing circuit removes any possibility as well for leaks at the junctions, as well for contamination of the trapping agent by over bubbling in the seawater sample. Presented data conﬁrm that this protocol allows recovering, with low variability between duplicate samples, as high as 99% of the 14CO2 produced during respiration experiments. Nevertheless, we demonstrated that eﬃciency of recovery varies between 77 and 99%. Such variability depends primarily on the extraction period, and on the volumetric ratio between sample and the extraction bottle. This means that this is a choice between a long extraction period oﬀering the best recovery
Figure 3. Linear regression between actually added and trapped 14CO2 radioactivity expressed as DPM (disintegrations per minute) per sample. Extraction was done at room temperature for 4 h with a stirring speed of 300 rpm, from 70 ml seawater samples distributed in 250 ml polycarbonate bottles (Nalgene). y ¼ 0.993 (p < 0.0001; n ¼ 42; R2 ¼ 0.996 with 0.973 < y < 1.014 with determination coeﬃcient of 95%).
6 eﬃciency, and a less eﬃcient, but shorter incubation period, because this last parameter cannot be neglected when processing large set of samples on board research vessels. Use of a multi-position electromagnetic stirrer may be helpful to found the best compromise between these two parameters. Indeed, to validate the measured respiration rates it is necessary to indicate precisely the used protocol, and to determine its actual eﬃciency. This particularly simple protocol optimizes and standardizes the recovery of 14CO2 produced during respiration assays of 14C-labeled organic compounds. Furthermore, variability remains particularly low, at least in the studied range from 1000 to 8000 DPM per sample, a range recovering most of the expected activities in oligotrophic pelagic seawaters. Because this protocol requires only widely available glassware, results should be reproducible with the same eﬃciency in any laboratory. So, when the planned respiratory assays are expected to provide activities entering in the above cited range of DPM, we suggest using one of the protocols described in Table 1, applying the corresponding eﬃciency coeﬃcient we already determined.
Acknowledgements This work was partially funded by the Institut National des Sciences de l’Univers and Total-FinaElf (Grant No. 11997 – CNRS/ENTREPRISE). C.T. was supported by a fellowship co-funded by INSU and Total-Fina-Elf. The authors thank A. Bianchi, J. Kuparinen, D. Kirchman and M. Yakimov for helpful comments on the manuscript.
References Bianchi, A., F. Van Wambeke & J. Garcin, 1998. Bacterial utilization of glucose in the water column from eutrophic to oligotrophic pelagic areas in the eastern North Atlantic Ocean. Journal of Marine System 14: 45–55. Bianchi, A., J. Garcin & O. Tholosan, 1999. Eﬀects of hydrostatic pressure on microbial activity through a 2000 m deep water column in the NW Mediterranean Sea. Marine Ecology Progress Series 183: 49–57. Dawson, R. & K. Gocke, 1978. Heterotrophic activity in comparison to the free amino acid concentrations in Baltic sea water samples. Oceanologica Acta 1: 45–54.
Deming, J. W., 1993. 14C tracer method for measuring microbial activity in deep-sea sediments. In Kemp, P. K., B. Sherr, E. Sherr & J. J. Cole (eds), Handbook of Methods in Aquatic Microbial Ecology. Lewis Publishers, Boca Raton, FL: 405–414. Foght, J. M., P. M. Fedorak & D. W. S. Westlake, 1989. Mineralization of [14C] hexadecane and [14C] phenanthrene in crude oil: speciﬁcity among bacterial isolates. Canadian Journal of Microbiology 36: 169–175. Garabe´tian, F., 1991. 14C-glucose uptake and 14CACO2 production in surface microlayer and surface-water samples: inﬂuence of UV and visible radiation. Marine Ecology Progress Series 77: 21–26. Hobbie, J. E. & C. C. Crawford, 1969. Respiration corrections for bacterial uptake of dissolved organic compounds in natural waters. Limnology & Oceanography 17: 725–730. Kirchman, D. L. & J. White, 1999. Hydrolysis and mineralization of chitin in the Delaware Estuary Aquatic Microbial Ecology 18: 187–196. Kuparinen, J. & A. Uusi-Rauva, 1980. A simpliﬁed technique to measure respiration rates of aerobic heterotrophic populations. Hydrobiologia 75: 113–115. Langdon, C., 1993. Community respiration measurements using a pulsed O2 electrode. In Kemp, P. K., B. Sherr, E. Sherr & J. J. Cole (eds), Handbook of Methods in Aquatic Microbial Ecology. Lewis Publishers, Boca Raton, FL: 447– 453. Nagata, T., R. Fukuda, I. Koike, K. Kogure & D. L. Kirchman, 1998. Degradation by bacteria of membrane and soluble protein in seawater. Aquatic Microbial Ecology 14: 29–37. Piehler, M. F., J. G. Swistak, J. L. Pinckney & H. W. Paerl, 1999. Stimulation of diesel fuel biodegradation by indigenous nitrogen ﬁxing bacterial consortia. Microbial Ecology 38: 69–78. Rooney-Varga, J. N., R. T. Anderson, J. L. Fraga, D. Ringelberg & D. R. Lovley, 1999. Microbial communities associated with anaerobic benzene degradation in a petroleum-contamined aquifer. Applied & Environmental Microbiology 65: 3056–3063. Seki, H., 1976. Method for estimating the decomposition of hexadecane in the marine environment. Applied and Environmental Microbiology 31: 439–441. Tamburini, C., J. Garcin & A. Bianchi, 2003. Role of deep-sea bacteria in organic matter mineralization and adaptation to hydrostatic pressure conditions in the NW Mediterranean Sea. Aquatic Microbial Ecology 32(3): 209–218. Unanue, M., I. Azua, J. M. Arrieta, A. Labirua-Iturburu, L. Egea & J. Iriberri, 1998. Bacterial colonization and ectoenzymatic activity in phytoplankton-derived model particles: cleavage of peptides and uptake of amino acids. Microbial Ecology 35(2): 103–112. Unanue, M., B. Ayo, D. Slezak, G. J. Herndl & J. Iberri, 1999. Ectoenzymatic activity and uptake of monomers in marine bacterioplankton described by a biphasic kinetic model. Microbial Ecology 37(1): 36–48. Walker, J. D. & R. R. Colwell, 1976. Measuring the potential activity of hydrocarbon-degrading bacteria. Applied & Environmental Microbiology 31: 189–197.
7 Williams, P. J. Le B., 1970. Heterotrophic utilization of dissolved organic compounds in the sea. I. Size distribution of population and relationship between respiration and incorporation of growth substrates. Journal of Marine Biologist Association, UK 50: 871–881. Williams, P. J. Le B., 1984. A review of measurements of respiration rates of marine plankton populations. In Hobbie, J.
& P. J. Le B. Williams (eds), Heterotrophic Activity in the Sea. Plenum Press, New York: 357–389. Williams, P. J. Le B. & C. Askew, 1968. A method of measuring the mineralization by micro-organisms of organic compounds in sea-water. Deep-Sea Research 15: 365–375. Wright, R. T. & J. E. Hobbie, 1965. The uptake of organic solutes in lake water. Limnology & Oceanography 10: 22–28.