Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Methane production by phosphate-starved SAR11 chemoheterotrophic marine bacteria

Methane production by phosphate-starved SAR11 chemoheterotrophic marine bacteria ARTICLE Received 9 Feb 2014 | Accepted 9 Jun 2014 | Published 7 Jul 2014 DOI: 10.1038/ncomms5346 Methane production by phosphate-starved SAR11 chemoheterotrophic marine bacteria 1, ,w 2, 1 1 Paul Carini * , Angelicque E. White *, Emily O. Campbell & Stephen J. Giovannoni The oxygenated surface waters of the world’s oceans are supersaturated with methane relative to the atmosphere, a phenomenon termed the ‘marine methane paradox’. The production of methylphosphonic acid (MPn) by marine archaea related to Nitrosopumilus maritimus and subsequent decomposition of MPn by phosphate-starved bacterioplankton may partially explain the excess methane in surface waters. Here we show that Pelagibacterales sp. strain HTCC7211, an isolate of the SAR11 clade of marine a-proteobacteria, produces methane from MPn, stoichiometric to phosphorus consumption, when starved for phosphate. Gene transcripts encoding phosphonate transport and hydrolysis proteins are upregulated under phosphate limitation, suggesting a genetic basis for the methanogenic phenotype. Strain HTCC7211 can also use 2-aminoethylphosphonate and assorted phosphate esters for phosphorus nutrition. Despite strain-specific differences in phosphorus utilization, these findings identify Pelagibacterales bacteria as a source of biogenic methane and further implicate phosphate starvation of chemoheterotrophic bacteria in the long-observed methane supersaturation in oxygenated waters. 1 2 Department of Microbiology, Oregon State University, Corvallis, Oregon 97331, USA. College of Earth, Ocean, and Atmospheric Sciences, Oregon State University, Corvallis, Oregon 97331, USA. * These authors contributed equally to this work. w Present address: University of Maryland Center for Environmental Sciences, Horn Point Laboratory, Cambridge, MD 21613, USA. Correspondence and requests for materials should be addressed to S.J.G. (email: [email protected]). NATURE COMMUNICATIONS | 5:4346 | DOI: 10.1038/ncomms5346 | www.nature.com/naturecommunications 1 & 2014 Macmillan Publishers Limited. All rights reserved. ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5346 he simplest of all hydrocarbons, methane (CH ), is a potent observations of CH maxima below the surface mixed layer and 4 4 greenhouse gas that has 420-fold more warming potential the lack of diurnal variation of these CH levels point to Tthan carbon dioxide, making an understanding of its heterotrophic organisms as probable contributors to CH sources and sinks important for global climate models (reviewed production—provided a steady source of MPn is available and in Valentine ). Canonical microbial methanogenesis pathways that environmental conditions favour the utilization of dissolved operate under strictly anaerobic conditions , such that organic phosphorus (DOP). Consistent with the hypothesis that methanogenesis is not favoured in marine surface waters where P-starved heterotrophic bacteria produce CH , members of oxygen is abundant. Historically, the only known sites of classical marine Vibrionales and Rhodobacterales expressed C–P lyase methanogenesis in the surface ocean were the anaerobic genes in P-limited mesocosm studies that, when amended with microenvironments of faecal matter and the guts of certain fish MPn, also generated CH (ref. 17). 3,4 or plankton . Yet, CH profiles in the open ocean show The SAR11 clade of oligotrophic a-proteobacteria (Pelagibac- subsurface maxima associated with the pycnocline (density, terales) are the numerically dominant chemoheterotrophic cells in s ¼ 25.2–25.9; 50–200 m) where CH is generally marine euphotic zones worldwide . Environmental and t 4 5,6 supersaturated in relationship to atmospheric equilibrium . laboratory studies have elucidated the forms of dissolved How this methane comes to be produced in oxygenic water organic matter Pelagibacterales bacteria use, including reduced masses is the basis of a phenomenon coined the ‘oceanic methane sulphur compounds, amino acids, one-carbon compounds, 7 29 30 paradox’ , which is particularly intriguing given that the isotopic organic acids (reviewed in Tripp ) and vitamin precursors . composition of subsurface CH maxima is inconsistent with that Despite a high degree of genomic conservation between 8 31 of anaerobic microenvironments . Pelagibacterales genomes , strategies for acquiring P appear to An alternate hypothesis has been proposed: CH is produced vary between Pelagibacterales isolates. For example, compared when marine microorganisms use methylphosphonic acid (MPn) with the northeast Pacific Ocean isolate, ‘Candidatus Pelagibacter as a resource for phosphorus (P) . Several lines of evidence ubique’ strain HTCC1062 (str. HTCC1062, herein), the genome support this ‘aerobic methane production’ hypothesis: (i) the C–P of Pelagibacterales sp. strain HTCC7211 (str. HTCC7211), lyase enzyme complex catabolizes MPn to CH , releasing isolated from the North Atlantic Ocean in the Sargasso Sea , 10,11 phosphate (P ) that can be used for growth (reviewed in encodes for additional proteins involved in organophosphorus 12 20 White and Metcalf ); (ii) C–P lyase gene sequences are present transport (PhnCDEE ) and phosphonate utilization 13–17 (sometimes abundant) in marine environments ; (iii) natural (PhnGHIJKLNM), suggesting that the two organisms have 9,17 seawater samples incubated with MPn release CH ; (iv) evolved different adaptive strategies for P acquisition related to biosynthesis of MPn by Nitrosopumilus maritimus SCM1, a niche partitioning . member of the ubiquitous and abundant Marine Group I (MGI) Here, building on these genomic insights we test the hypothesis archaea was recently demonstrated ; and (v) ca. 0.6% of that Pelagibacterales chemoheterotrophic bacteria encoding the microbes in marine surface communities contain genes C–P lyase complex produce CH from MPn when P starved. We 4 i encoding the MPn synthase . Taken together, these findings show that str. HTCC7211 produces CH from MPn in suggest the observed CH supersaturation can be explained by the proportions stoichiometric to P demand. Moreover, we present natural synthesis of MPn and subsequent hydrolysis for purposes rates of CH production under ideal growth conditions, together of P nutrition by marine plankton. with the genetic response to P starvation in the Pelagibacterales. For this process to act as a substantial source of CH , there These data suggest Pelagibacterales are one of the leading must be an ecophysiological niche for MPn utilization in the contributors to CH dynamics in P -deficient oligotrophic gyres. 4 i oxygenic surface ocean. Isolated cultures of Escherichia coli, 10,12 Pseudomonas spp. and other heterotrophic bacteria , as well as marine autotrophs such as Trichodesmium , have been shown to Results use MPn as a sole P source. In these organisms, MPn degradation The capacity of Pelagibacterales strains HTCC1062 and proceeds via the C–P lyase pathway and is induced by HTCC7211 to grow on multiple alternate P sources, including 12,20–23 P starvation . Building on these findings, studies phosphite, phosphonates and phosphate esters, was tested. Strain 9 17 by Karl et al. and Martınez et al. showed that when surface HTCC1062 grew on P and did not use added DOP (Fig. 1). In seawater is amended with glucose and a nitrogen source contrast, str. HTCC7211 used a broad range of alternate (to stimulate P -limiting conditions), MPn hydrolysis and CH compounds for P nutrition, including P esters (phosphoserine, i 4 production proceeds at rates that scale inversely with the glucose 6-phosphate and ribose 5-phosphate), phosphonates addition of exogenous P . The authors implicate a combination (2-aminoethylphosphonic acid and MPn) and reduced inorganic of heterotrophic marine bacteria, together with natural P (phosphite) (Fig. 1). In all cases, diauxic growth was observed assemblages of Trichodesmium spp., as drivers of this when str. HTCC7211 cells were grown with an alternate P source 9,17 metabolism . These findings are consistent with studies of (Fig. 2); this growth pattern is presumably related to a transition 23 24 gene expression and methane production in Trichodesmium from growth on P to growth on organic P resources as has been 33,34 erythraeum IMS101 that show induction of C–P lyase genes described for other organisms . There was no significant under P starvation and subsequent methane production from difference in negative control (no P added) cell yields and the cell MPn. yields attained when glucose 6-phosphate, ribose 5-phosphate, Although these experiments provide definitive evidence for MPn, 2-AEP and phosphoserine were supplied to str. HTCC1062 potential MPn utilization in the surface ocean, they do not (Fig. 1). We presume that this is an indication that abiotic P necessarily address the CH supersaturation observed at the release from these compounds is negligible. The increased str. pycnocline, nor the organism(s) responsible for production of HTCC1062 cell yields when grown with phosphite as a sole excess CH in these deep layers of the euphotic zone. It seems source of P could result from (i) trace amounts of phosphate; (ii) unlikely that the activity of large buoyant genera such as abiotic conversion (oxidation) of phosphite to phosphate; or (iii) Trichodesmium can wholly explain the subsurface CH features the ability of str. HTCC1062 to enzymatically oxidize phosphite given that Trichodesmium are largely observed within or at the to phosphate via an unknown mechanism. At this time we do not 25,26 base of surface mixed layer and that CH supersaturation has have evidence that supports one of these possibilities over the also been observed outside of the subtropical gyres . Alternately, other. 2 NATURE COMMUNICATIONS | 5:4346 | DOI: 10.1038/ncomms5346 | www.nature.com/naturecommunications & 2014 Macmillan Publishers Limited. All rights reserved. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5346 ARTICLE str.HTCC1062 str. HTCC7211 No P (– control) P (+ control) MPn Phosphite 2-AEP Ribose-5P Glucose-6P P-serine 5 101520253035 Time (days) Figure 2 | Diauxic growth of Pelagibacterales sp. str. HTCC7211 on alternate phosphorus sources. Points are the mean densities of biological replicates 1.0 s.d. (n¼ 3). Each alternate phosphorus compound was Figure 1 | P-source utilization by Pelagibacterales strains HTCC1062 supplied at a final concentration of 1.0mM. P was supplied as NaH PO i 2 4 (blue) and HTCC7211 (red). Bar height is the mean of the maximum density in the positive control. Growth medium was not amended with P in the achieved for biological replicates 1.0 s.d. (n¼ 3). P was supplied as negative control. Remaining treatments had 1.0mM of P source added. NaH PO in the positive control. Growth medium was not amended with P in 2 4 i Cell inoculum was previously grown on P . the negative control. Remaining treatments had 1.0mM of P source added. The genetic basis for the P source utilization patterns observed metaproteomes , suggesting an important role in the response for each strain was investigated at a transcriptional level using to P stress. Given its genomic context adjacent to an DNA microarrays. Time-course experiments showed that organophosphate transporter (phnCDEE ) and strong P -starved str. HTCC1062 cultures rapidly induced expression upregulation under P -deplete conditions, we postulate that i i of pstSCAB, encoding the high-affinity transport system (Fig. 3). PB7211_828 may be involved in P-ester hydrolysis. Individual pst genes were upregulated 2.4- to 3.6-fold 4 h after the CH is a product of MPn hydrolysis by P -starved str. 4 i onset of P starvation and 9.1- to 27.9-fold 38 h after the onset of HTCC7211 cells. When synthetic growth medium (without P starvation (Fig. 3 and Supplementary Tables 1–3). A added P ) was amended with excess MPn (10mM), str. i i concomitant upregulation of global stress-response genes (includ- HTCC7211 produced CH (Fig. 4a) concomitantly with cell ing recA, lexA and umuD) and downregulation of ribosomal growth (Fig. 4b). CH was not produced when cells were grown protein transcripts (including rpsB, rpsP, rpsK, rpsS, rplB and with P additions (Fig. 4a) and CH production from MPn was i 4 rplC) suggested that P depletion is treated as a global stressor in repressed when cells were grown in the presence of MPn and str. HTCC1062 (Supplementary Tables 2–3). P starvation of str. P (Table 1). Mass balance confirmed this observation: i i ± ± HTCC7211 induced expression of genes encoding the organo- 1.29 0.18mMCH (mean s.d.; n¼ 3; presumably from the phosphorus ABC transporter (phnCDEE ; 7.3- to 21.8-fold) and catabolism of 1.29 0.18mM MPn) was produced over the course the C–P lyase (phnGHIJKLM; 2.1- to 7.3-fold) (Fig. 3 and of the 36-day incubation, plus 8.47 0.21mM residual DOP Supplementary Tables 4 and 5). Genes encoding str. HTCC7211’s (mean s.d.; n¼ 3) measured at the end of the experiment, high-affinity P transporter pstSCAB were also upregulated, but to account for the full complement of added MPn (10mM). In a lesser degree (3.2- to 3.7-fold) than the organophosphorus culture, cell-normalized CH production was 15.2 0.04 amol transporter. In contrast to str. HTCC1062, stress-response genes CH per cell (mean s.d.; n¼ 3) at an average rate of 0.42 amol were not differentially regulated in str. HTCC7211. CH per cell per day. For comparison, the P quota of MPn-grown Although str. HTCC7211 used the P-esters glucose 6-phos- str. HTCC7211 was 10.04 0.02 amol per cell (Table 2), phate, ribose 5-phosphate and phosphoserine for growth, the suggesting that the hydrolysis of MPn exceeded the cellular P mechanism of utilization is unclear and might involve a novel demand under these ideal growth conditions. At high P source phosphatase. Genes encoding recognized bacterial alkaline concentrations (467 nM), significant growth rate differences phosphatases, commonly associated with P-ester hydrolysis were observed between the MPn and P treatments (0.29 0.00 (phoA, phoD and phoX) , are absent from the str. HTCC7211 and 0.36 0.01 per day, respectively; t-test P-value: 0.000007, genome. We speculate that str. HTCC7211 encodes a novel n¼ 6) (Table 3). Overall growth rates were slower at lower phosphatase, with a broad substrate range, that may explain the concentrations (o67 nM) of MPn or P but did not differ ± ± observed utilization of P esters. The HD-hydrolase (‘HD’ in significantly as a function of P source (0.22 0.02 and 0.24 0.04 Fig. 3b), induced by P depletion, is a member of a large family of per day, respectively; t-test P-value: 0.35, n¼ 6) (Table 3). metal-dependent phosphohydrolases and may act on P esters to release P . Interestingly, a second gene (PB7211_828; ‘828’ in Fig. 3b), divergently transcribed downstream of phnE Discussion and annotated as a ‘hypothetical protein’, was upregulated Our data reveal a differential capacity for DOP utilization among 12- to 15-fold in P -deplete conditions. Although PB7211_828 the Pelagibacterales, with str. HTCC7211, but not str. HTCC1062, is unique to str. HTCC7211 (a so-called ORFan ), peptides having a broad enzymatic capacity for DOP utilization (Fig. 1). mapping to PB7211_828 were identified in Sargasso Sea MPn was used by str. HTCC7211 in place of P (Fig. 1) and when NATURE COMMUNICATIONS | 5:4346 | DOI: 10.1038/ncomms5346 | www.nature.com/naturecommunications 3 & 2014 Macmillan Publishers Limited. All rights reserved. –P (neg. ctr) Phosphite Glucose-5P Ribose-5P MPn 2-AEP P-serine +P (pos. ctr) 10 –1 Max. density (× 10 cells L ) –1 Density (cells L ) ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5346 Ca.P. ubique str. HTCC1062 High-affinity P transport phoB phoU pstB pstA pstC pstS phoR 1181 4 h ≥30.0 20 h 2.0 Amount upregulated (fold) 38 h b Pelagibacterales sp. str. HTCC7211 High-affinity P transport Phosphonate transport C-P lyase pstS pstC pstA pstB phoU phoB HD 828 phnE2 phnE phnD phnC phnF G H phnI phnJ phnK phnL N 764 phnM 68 h 96 h Figure 3 | Comparative genomics of select P acquisition-related gene expression in P -starved cell suspensions of Pelagibacterales strains HTCC1062 (a) and HTCC7211 (b). The time elapsed since the onset of resuspension is shown to the left of each gene row. Vertical lines between genes in b represent a lack of genomic synteny. Genes shaded grey were not significantly differentially expressed. Gene names, annotations and actual expression values are listed in Supplementary Tables 1–5. 1181, SAR11_1181; HD, H-D Hydrolase; 828, PB7211_828; G, phnG;H, phnH;N, phnN, 764, PB7211_764. doing so, CH was stoichiometrically released (Fig. 4). Consistent above. In situ CH production rates are probably a function of (i) 4 4 with these observations, genes encoding enzymes and transpor- MPn supply relative to the total bioavailable DOP pool accessible ters necessary to catabolize MPn (and other phosphonates) were to the Pelagibacterales; (ii) the relative efficiency with which MPn upregulated under P limitation (Fig. 3). MPn hydrolysis was is used when multiple substrates are simultaneously available; and strongly repressed by P (Table 1), suggesting that in situ CH (iii) the percent of the Pelagibacterales population that contain C– i 4 production via MPn degradation by Pelagibacterales bacteria is P lyase genes. We speculate that maximal CH production by likely to be confined to P -deplete ocean regimes, such as the Pelagibacterales may occur shortly after the seasonal spring Sargasso Sea. Accordingly, published metagenomic analyses of mixing event in the Sargasso Sea. This hypothesis is consistent multiple oceanic environments show that the Pelagibacterales-like with the observation that MPn producers (ammonia-oxidizing C–P lyase genes are not universally distributed, but instead are MGI archaea) are most active at 120 m, the base of the euphotic 13–15 42 more abundant in P -deplete waters . zone in the Sargasso Sea , and that Pelagibacterales stocks Although Pelagibacterales are likely to contribute meaningfully increase sharply after spring mixing . to the observed CH supersaturation in surface waters because of The laboratory results presented here help to contextualize 13–15 38 their large population size, calculating their specific contribution metagenomic and metaproteomic findings that suggested is challenging owing to the uncertainties pertaining to that Pelagibacterales bacteria actively transport and use DOP, strain heterogeneity with respect to C–P lyase gene content and including phosphonates and P-esters, in the Sargasso Sea (Fig. 1). how that may relate to the observed niche partitioning of The significance of these findings is extended to the global climate Pelagibacterales ecotypes. In the Sargasso Sea, total Pelagibacter- by showing that P -starved Pelagibacterales produce CH when i 4 8  1 ales cell abundances oscillate between 0.5 and 2.2 10 cells l MPn is provided as a sole P source (Fig. 4). Interestingly, in with maximum abundances at or above the pycnocline . addition to the abundant MGI archaea, some Pelagibacterales Spatiotemporal variation in the distribution of Pelagibacterales ecotypes may also harbour genes encoding the MPn synthase 31,40,41 18 ecotypes has been discussed in multiple studies . The degree enzyme , suggesting Pelagibacterales may also be an MPn source to which the presence of C–P lyase genes correlates with the in the ocean. An understanding of environmental MPn dynamics different Pelagibacterales ecotypes and whether the relative (including supply and consumption rates) as it pertains to DOP abundances of Pelagibacterales-like C–P lyase genes fluctuate utilization, remains a significant challenge to resolving CH fluxes seasonally is unknown. When analysing pyrosequences from as they pertain to P stress in marine environments. the Sargasso Sea, Martinez et al. showed that 82.5% of Although methane production by P -starved bacterioplankton the identifiable phnI genes (used as a proxy for the presence is a probable source of CH in P -deplete waters rich in MPn, 4 i of the C–P lyase) best hit to str. HTCC7211’s phnI gene, there may be other CH -producing biological pathways that bear suggesting that Pelagibacterales members mediate a large portion consideration. For example, in the nitrate-deplete, P -replete of the C–P lyase activity in the Sargasso Sea. Using a metric called waters of the central Arctic Ocean, an inverse correlation ‘multiplicity per cell,’ Coleman and Chisholm analysed between CH and a common resource for bacterial carbon and metagenomic data from the Sargasso Sea (collected in October sulphur, dimethylsulphoniopropionate (DMSP), was found .In of 2006) and calculated that B52% of all Pelagibacterales cells subsequent incubation experiments, when DMSP was added contained phnJ, a marker gene used to infer the presence of the to the Arctic surface waters, CH was produced and a- and 16 27 C–P lyase gene suite . Multiplying the approximate abundance g-proteobacteria increased in relative abundance . From these of C–P lyase containing Pelagibacterales cells found in results, Damm et al. hypothesized a type of ‘methylotrophic the P-deplete Sargasso Sea (52%, based on phnJ,of methanogenesis’ in which the methylated degradation products of 8  1 0.5–2.2 10 cells l ) by the maximal net cell-specific CH DMSP (methanethiol in particular) act as a precursor for CH 4 4 production rate measured here (0.42 amol CH per cell per day; production. Although this is an intriguing possibility that may Fig. 4), we estimate maximal CH production potential of ca. represent an alternate, and potentially significant, CH source in 4 4 0.01–0.05 nM per day via MPn hydrolysis. aerobic surface waters, the enzymes, pathways and organisms As MPn is part of a complex milieu of dissolved organopho- involved have not yet been identified. Further study of sphorus compounds that Pelagibacterales can use (for example, mechanisms of aerobic CH production in relationship to those illustrated in Fig. 1), the actual CH production rates are bacterial diversity and gene expression are needed to further probably much lower than the potential contribution calculated unravel oceanic CH sources. 4 NATURE COMMUNICATIONS | 5:4346 | DOI: 10.1038/ncomms5346 | www.nature.com/naturecommunications & 2014 Macmillan Publishers Limited. All rights reserved. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5346 ARTICLE Table 1 | CH production by str. HTCC7211 from MPn (10 lM) as a function of P addition*. 1.4 P added (lM) Net CH produced (lM) % Repressionw i 4 1.2 0 1.29 0 0.08 0.76 41 1.0 0.125 0.14 89 0.25 0.05 96 0.8 0.5 0.05 96 0.75 0.04 97 0.6 1 0.04 97 i 5 0.03 97 0.4 10 0.02 98 MPn 0.2 MPn, methylphosphonic acid. *Values after 36-day incubation. 0 w% Repression¼ 100 (1 (net CH produced/1.29)). Table 2 | Pelagibacterales sp. str. HTCC7211 cell yield as a function of added P. P treatment MPn treatment Added P , nM Maximum Added MPn, nM Maximum 1  1 cell yield (l )* cell yield (l )* 9 9 ± ± 98 9.41 0.36 10 67 7.31 0.23 10 9 9 ± ± i 9.8 1.21 0.14 10 6.7 1.30 0.05 10 8 8 ± ± MPn 0.98 6.78 1.53 10 0.67 7.14 0.23 10 No P ± ± i P quotaw 10.95 0.30 amol P quotaw 10.04 0.02 amol 1  1 1 P cell P cell MPn, methylphosphonic acid. *Cell yields are presented as the mean s.d. (n¼ 3). wThe P quota of str. HTCC7211 cells was determined via a dose–response curve using cultures grown on P or MPn. Triplicate cultures were created for each P dose and grown under the conditions described in ‘cultivation methods’ until P-limited stationary phase was reached. 0.12 The P quota was derived from linear regression s.d. through the maximum cell yield versus the initial dissolved P added. 0.10 0.08 Table 3 | Specific growth rates of str. HTCC7211 in AMS1 0.06 with varying levels of P or MPn. 0.04 P treatment MPn treatment CH 4 Added Specific growth rate Added Specific growth rate 0.02 2 P,nM (per day)* MPn, nM (per day)* Cell i ± ± 9800 0.35 0.01 6700 0.29 0.00 ± ± 98 0.36 0.01 67 0.29 0.01 0 ± ± 9.8 0.27 0.01 6.7 0.24 0.01 0 5 10 15 20 25 30 35 40 ± ± 0.98 0.20 0.02 0.67 0.21 0.01 Days ± ± Mean 0.29 0.07 Mean 0.26 0.04 Figure 4 | Pelagibacterales str. HTCC7211 produces methane when P MPn, methylphosphonic acid. starved. (a) Total CH concentration in sealed bottles from Pelagibacterales ± *Specific growth rates are presented as the mean s.d. (n¼ 3). sp. str. HTCC7211 cultures grown with MPn or P . Points are the mean CH concentration 1.0 s.d. (n¼ 3) (b) Cell densities from bottles of 4nM; B : 700 pM; Myo-inositol: 6mM; 4-aminobenzoate: 60 nM). P sources were Pelagibacterales sp. str. HTCC7211 cultures grown with no P , MPn or P.Cell i i added as indicated through the main text. counts for each treatment were obtained from single bottles used solely for cell counts. (c) The CH production rate (change in CH concentration per 4 4 day) and cell production rate (change in cell density per day) from MPn Cultivation details. All cultures, except those describing CH production (described below), were grown in acid-washed and autoclaved polycarbonate flasks. treatments before day 25, after which cell concentrations began to decline. Cultures were incubated at 20 C with shaking at 60 r.p.m. under a 12-h light:12-h dark cycle. Light levels during the day were held at 140–180mmol photons 2  1 m s . Cells for cell counts were stained with SYBR green I and counted with a Methods Guava Technologies flow cytometer at 48–72 h intervals as described elsewhere . Organism source. ‘Candidatus Pelagibacter ubique’ str. HTCC1062 (ref. 43) and The treatments for the alternate phosphate source utilization experiments were: no Pelagibacterales sp. str. HTCC7211 (ref. 32) were revived from 10% glycerol stocks P addition (negative control), 1.0mMP (as NaH PO ; positive control), 1.0mM i i 2 4 and propagated in artificial seawater medium for SAR11 (AMS1) , without added phosphite (as NaPHO 5H O), 1.0mM glucose 6-phosphate, 1.0mM ribose 3 2 P , but amended with pyruvate (100mM), glycine (5mM), methionine (5mM), FeCl 5-phosphate, 1.0mM MPn, 1.0mM 2-aminoethylphosphonic acid or 1.0 mM i 3 (1mM), and vitamins (B :6mM; B : 800 nM; B : 425 nM; B : 500 nM; B :4nM;B : phosphoserine. All alternate phosphorus compounds were Z98% pure. 1 3 5 6 7 9 NATURE COMMUNICATIONS | 5:4346 | DOI: 10.1038/ncomms5346 | www.nature.com/naturecommunications 5 & 2014 Macmillan Publishers Limited. All rights reserved. –1 CH production rate (nM d ) –1 10 Total CH (μM) Density (cells L × 10 ) –1 –1 9 Cell production rate (L d × 10 ) ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5346 Methane production from MPn. Pelagibacterales sp. str. HTCC7211 cells, pre- References viously grown in media with MPn as the sole P source, were used to inoculate 1. Valentine, D. L. Emerging topics in marine methane biogeochemistry. Annu. sterile, BSA-coated glass serum bottles containing growth media amended with Rev. Marine. Sci. 3, 147–171 (2011). 10mM MPn as the sole P source. The negative control consisted of str. HTCC7211 2. Reeburgh, W. S. Oceanic methane biogeochemistry. Chem. Rev. 107, 486–513 cells, previously grown in media with P as the sole P source, as inoculum for i (2007). sterile, BSA-coated serum bottles containing growth media amended with 10mMP 3. Tilbrook, B. D. & Karl, D. M. Methane sources, distributions and sinks from as the sole P source. Serum bottles (60 ml) were filled with 55 ml culture, leaving a California coastal waters to the oligotrophic North Pacific gyre. Mar. Chem. 49, 5-ml headspace, capped with Viton septa, crimped with aluminum seals and 51–64 (1995). incubated horizontally in the conditions described in ‘Cultivation details.’ Cell 4. Sasakawa, M. et al. Carbon isotopic characterization for the origin of excess counts for each treatment were obtained from single bottles used solely for methane in subsurface seawater. J. Geophys. Res. 113, C03012 (2008). cell counts. 5. Burke, R. J., Reid, D. F., Brooks, J. M. & Lavoie, D. M. Upper water column CH in the headspace was measured by gas chromatography using a Shimadzu methane geochemistry in the Eastern Tropical North Pacific. Limnol. Oceanogr. GC-8A gas chromatograph equipped with a column packed with Porapak N 28, 19–32 (1983). (80/100 mesh size) fitted with a flame ionization detector. In brief, 100ml samples 6. Holmes, M. E., Sansone, F. J., Rust, T. M. & Popp, B. N. Methane production, of headspace from each bottle were injected into the GC at a flow rate of 1 consumption, and air-sea exchange in the open ocean: An Evaluation based on 25 ml min using ultrapure helium (He) as the carrier gas. Peaks were integrated carbon isotopic ratios. Global Biogeochem. Cycles 14, 1–10 (2000). using PeakSimple chromatography software (http://www.srigc.com/). 7. Kiene, R. P. in Microbial Production and Consumption of Greenhouse Gases Quantification of CH was accomplished by calibrating peak areas to the flame Methane, Nitrogen Oxides, and Halomethanes 111–146 (ASM, 1991). ionization detector response to a three-point calibration conducted at the beginning and end of the experiment using 100ml injections of 100, 1,000 and 8. Sansone, F. J., Graham, A. W. & Berelson, W. M. Methane along the western 10,000 p.p.m. CH standards. The 1,000 and 10,000 p.p.m. standards were prepared Mexican margin. Limnol. Oceanogr. 49, 2242–2255 (2004). from a 1% mixed gas standard containing CH , ethane and ethylene. Replicates of 4 9. Karl, D. M. et al. Aerobic production of methane in the sea. Nat. Geosci. 1, the 100 p.p.m. standard were analysed alongside samples on each sampling day. 473–478 (2008). Over the course of the experiment, the peak area for the 100 p.p.m. standard varied 10. Daughton, C. G., Cook, A. M. & Alexander, M. Biodegradation of phosphonate by o5% (coefficient of variation). toxicants yields methane or ethane on cleavage of the C-P bond. FEMS Headspace equilibration models at 20 C using a Bunsen coefficient of 0.0345 Microbiol. Lett. 5, 91–93 (1979). for a 60 ml total volume (55 ml liquid, 5 ml headspace) suggest the fraction of the 11. Kamat, S. S., Williams, H. J. & Raushel, F. M. Intermediates in the methane in the gas phase to be 74.5%. The fractionation of CH between headspace transformation of phosphonates to phosphate by bacteria. Nature 480, 570–573 and aqueous phase was verified by injecting commercial CH standards (1.0 ml of (2011). either 100 p.p.m. or 10,000 p.p.m. CH equivalent to 4.08 nmol CH , 40.8 nmol 4 4 12. White, A. K. & Metcalf, W. W. Microbial metabolism of reduced phosphorus CH and 408 nmol CH , respectively) into sealed serum bottles containing 4 4 compounds. Annu. Rev. Microbiol. 61, 379–400 (2007). uninoculated growth medium. Of the added CH ,72 6% in these standards was 13. Martinez, A., Tyson, G. W. & Delong, E. F. Widespread known and novel measured in the headspace at 20 C. A value of 74% was used to adjust headspace phosphonate utilization pathways in marine bacteria revealed by functional CH measurements to total CH . 4 4 screening and metagenomic analyses. Environ. Microbiol. 12, 222–238 (2010). 14. Coleman, M. L. & Chisholm, S. W. Ecosystem-specific selection pressures revealed through comparative population genomics. Proc. Natl Acad. Sci. 107, Dissolved phosphorus measurements. Samples for determination of total 18634–18639 (2010). dissolved P levels were obtained at the end of CH bioassays (see ‘Methane pro- 15. Martiny, A. C., Huang, Y. & Li, W. in Handbook of Molecular Microbial Ecology duction from MPn’). Cells were removed from spent growth medium by filtration II: Metagenomics in Different Habitats 269–276 (John Wiley & Sons, Inc., through a 0.02-mm Acrodisc filter. The dissolved inorganic P (DIP) concentration of the filtrate was determined colorimetrically as described by Murphy and Riley 2011). using a Cary ultraviolet–visible spectrophotometer. The total dissolved P 16. Villarreal-Chiu, J. F., Quinn, J. P. & Mcgrath, J. W. The genes and enzymes of concentration of the filtrate was measured using a modification of the phosphonate metabolism by bacteria, and their distribution in the marine Valderrama method . Briefly, samples were oxidized at 120 C for 60 min in environment. Front. Microbiol. 3, 19–19 (2012). Teflon digestion bombs, allowed to cool and run as DIP, as described above. 17. Martınez, A., Ventouras, L. A., Wilson, S. T., Karl, D. M. & DeLong, E. F. The oxidizing agent used was a solution of potassium persulphate, sodium Metatranscriptomic and functional metagenomic analysis of hydroxide and boric acid (as per ref. 46) added in a 1:10 oxidant to sample methylphosphonate utilization by marine bacteria. Front. Microbiol. 4, 340 ratio. DOP concentrations were calculated as the difference between total dissolved (2013). P and DIP. 18. Metcalf, W. W. et al. Synthesis of methylphosphonic acid by marine microbes: a source for methane in the aerobic ocean. Science (New York, NY) 337, 1104–1107 (2012). Microarray growth and sampling conditions. Biological replicates (n¼ 3) of str. 19. White, A. E., Karl, D. M., Bjo¨rkman, K., Beversdorf, L. J. & Letelier, R. M. HTCC1062 or str. HTCC7211 were grown in AMS1 (ref.39) amended with Production of organic matter by Trichodesmium IMS101 as a function of pyruvate (1 mM), glycine (50mM), methionine (50mM), FeCl (1mM), P (10mM) 3 i phosphorus source. Limnol. Oceanogr. 55, 1755–1767 (2010). and vitamins (as described above). In mid- to late-logarithmic growth (ca. 0.7- 8  1 20. Metcalf, W. W. & Wanner, B. L. Involvement of the Escherichia coli phn (psiD) 1.0 10 cells ml ), cells were harvested by centrifugation (17,664g for 1.0 h at gene cluster in assimilation of phosphorus in the form of phosphonates, 20 C). Cell pellets were evenly split and washed twice with growth media that was phosphite, Pi esters, and Pi. J. Bacteriol. 173, 587–600 (1991). either amended with 100mMP (replete conditions) or not amended with P i i 21. Wanner, B. L. & Metcalf, W. W. Molecular genetic studies of a 10.9-kb operon (deplete conditions). After washing, pellets were re-suspended in P -deplete or in Escherichia coli for phosphonate uptake and biodegradation. FEMS P -replete growth media. Samples were collected from cell suspensions at t¼ 0, 4, 20 and 38 h after resuspension for str. HTCC1062, and t¼ 0, 20, 38, 68 and 96 h Microbiol. Lett. 100, 133–139 (1992). after resuspension for str. HTCC7211. Sampling consisted of centrifugation 22. Metcalf, W. W. & Wanner, B. L. Mutational analysis of an Escherichia coli (48,298g for 1.0 h at 4 C) followed by resuspension of cell pellets in ‘RNAprotect fourteen-gene operon for phosphonate degradation, using TnphoA’ elements. Bacteria’ reagent (Qiagen, Valencia, CA). RNA was extracted using an RNeasy J. Bacteriol. 175, 3430–3442 (1993). Mini kit (Qiagen) and amplified using the MessageAmp-II Bacteria RNA 23. Dyhrman, S. et al. Phosphonate utilization by the globally important marine amplification kit (Ambion, Carlsbad, CA) using the manufacturer’s instructions. diazotroph Trichodesmium. Nature 439, 68–71 (2006). Amplified and labelled RNA was hybridized to custom ‘Candidatus Pelagibacter 24. Beversdorf, L. J., White, A. E., Bjo¨rkman, K. M., Letelier, R. M. & Karl, D. M. ubique’ Affymetrix GeneChip arrays (Pubiquea520471f) (Affymetrix, Santa Clara, Phosphonate metabolism by Trichodesmium IMS101 and the production of CA) that contained probes for Pelagibacterales strains HTCC1002, HTCC1062 and greenhouse gases. Limnol. Oceanogr. 55, 1768–1778 (2010). HTCC7211 as described previously . Briefly, Affymetrix GeneChip arrays were 25. Davis, C. S. & McGillicuddy, D. J. Transatlantic abundance of the N2-fixing hybridized and scanned using an Affymetrix GeneChip fluidics station, colonial cyanobacterium Trichodesmium. Science (New York, NY) 312, 1517– hybridization oven and scanner. Background and raw expression values were 1520 (2006). normalized between chips using the robust microarray algorithm using the 26. Letelier, R. M. & Karl, D. M. Role of Trichodesmium spp. in the productivity of Affymetrix Expression Console software. A Bayesian statistical analysis was 48 the subtropical North Pacific Ocean. Mar. Ecol. Prog. Ser. 133, 263–273 (1996). conducted using Cyber-T . The estimate of variance was calculated in 27. Damm, E. et al. Methane production in aerobic oligotrophic surface water in Cyber-T by using window sizes of 101 and a confidence value of 10. A t-test was the central Arctic Ocean. Biogeosciences 7, 1099–1108 (2010). performed on log-transformed expression values by using the Bayesian variance 28. Morris, R. et al. SAR11 clade dominates ocean surface bacterioplankton estimate. The programme QVALUE, was used to obtain a q-value, which accounts for multiple t-tests performed . A gene was defined as differentially expressed communities. Nature 420, 806–810 (2002). 29. Tripp, H. J. The unique metabolism of SAR11 aquatic bacteria. J. Microbiol. 51, if both the q-value was r0.05 and the gene was differentially regulated by Z2.0-fold. 147–153 (2013). 6 NATURE COMMUNICATIONS | 5:4346 | DOI: 10.1038/ncomms5346 | www.nature.com/naturecommunications & 2014 Macmillan Publishers Limited. All rights reserved. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5346 ARTICLE 30. Carini, P. et al. Discovery of a SAR11 growth requirement for thiamin’s 45. Murphy, J. & Riley, J. P. A modified single solution method for the pyrimidine precursor and its distribution in the Sargasso Sea. ISME J. determination of phosphate in natural waters. Anal. Chim. Acta 27, 31–36 (1962). doi:10.1038/ismej.2014.61. 46. Valderrama, J. C. The simultaneous analysis of total nitrogen and total 31. Grote, J. et al. Streamlining and core genome conservation among highly phosphorus in natural waters. Mar. Chem. 10, 109–122 (1981). divergent members of the SAR11 clade. MBio 3, e00252–12 (2012). 47. Schwalbach, M. S., Tripp, H. J., Steindler, L., Smith, D. P. & Giovannoni, S. J. 32. Stingl, U., Tripp, H. J. & Giovannoni, S. J. Improvements of high-throughput The presence of the glycolysis operon in SAR11 genomes is positively correlated culturing yielded novel SAR11 strains and other abundant marine bacteria from with ocean productivity. Environ. Microbiol. 12, 490–500 (2010). the Oregon coast and the Bermuda Atlantic Time Series study site. ISME J. 1, 48. Baldi, P. & Long, A. D. A Bayesian framework for the analysis of microarray 361–371 (2007). expression data: regularized t -test and statistical inferences of gene changes. 33. Daughton, C. G., Cook, A. M. & Alexander, M. Phosphate and soil binding: Bioinformatics 17, 509–519 (2001). factors limiting bacterial degradation of ionic phosphorus-containing pesticide 49. Storey, J. D. & Tibshirani, R. Statistical significance for genomewide studies. metabolites. Appl. Environ. Microbiol. 37, 605–609 (1979). Proc. Natl Acad. Sci. 100, 9440–9445 (2003). 34. Wackett, L. P., Wanner, B. L., Venditti, C. P. & Walsh, C. T. Involvement of the phosphate regulon and the psiD locus in carbon-phosphorus lyase activity of Escherichia coli K-12. J. Bacteriol. 169, 1753–1756 (1987). Acknowledgements 35. Luo, H., Benner, R., Long, R. A. & Hu, J. Subcellular localization of We thank Katie Watkins-Brandt for P measurements and Kiriann L. Carini for con- ceptual thumbnail. Microarrays were hybridized at the Oregon State University Center marine bacterial alkaline phosphatases. Proc. Natl Acad. Sci. 106, 21219–21223 for Genome Research and Biocomputing Core Laboratory. This research was funded by (2009). 36. Aravind, L. & Koonin, E. V. The HD domain defines a new superfamily of the Gordon and Betty Moore Foundation through Grant GBMF607.01 to Stephen Giovannoni and National Science Foundation grants OCE-0802004 and OCE-0962362. metal-dependent phosphohydrolases. Trends Biochem. Sci. 23, 469–472 (1998). 37. Fischer, D. & Eisenberg, D. Finding families for genomic ORFans. Bioinformatics 15, 759–762 (1999). Author contributions 38. Sowell, S. M. et al. Transport functions dominate the SAR11 metaproteome at P.C., A.W. and S.J.G. designed experiments and prepared manuscript. P.C., A.W. and low-nutrient extremes in the Sargasso Sea. ISME J. 3, 93–105 (2009). E.O.C. conducted experiments. 39. Carini, P., Steindler, L., Beszteri, S. & Giovannoni, S. J. Nutrient requirements for growth of the extreme oligotroph ‘Candidatus Pelagibacter ubique’ Additional information HTCC1062 on a defined medium. ISME J. 7, 592–602 (2013). Accession codes: The raw microarray data files are deposited in the NCBI Gene 40. Carlson, C. A. et al. Seasonal dynamics of SAR11 populations in the euphotic Expression Omnibus (GEO) database with accession codes GSM1318808 to and mesopelagic zones of the northwestern Sargasso Sea. ISME J. 3, 283–295 GSM1318861. (2009). 41. Vergin, K. L. et al. High-resolution SAR11 ecotype dynamics at the Bermuda Supplementary Information accompanies this paper at http://www.nature.com/ Atlantic Time-series Study site by phylogenetic placement of pyrosequences. naturecommunications ISME J. 7, 1322–1332 (2013). Competing financial interests: The authors declare no competing financial interests. 42. Newell, S. E., Fawcett, S. E. & Ward, B. B. Depth distribution of ammonia oxidation rates and ammonia-oxidizer community composition in the Sargasso Reprints and permission information is available online at http://npg.nature.com/ Sea. Limnol. Oceanogr 58, 1491–1500 (2013). reprintsandpermissions/ 43. Rappe´, M. S., Connon, S. A., Vergin, K. L. & Giovannoni, S. J. Cultivation of the ubiquitous SAR11 marine bacterioplankton clade. Nature 418, 630–633 (2002). How to cite this article: Carini, P. et al. Methane production by phosphate-starved 44. Tripp, H. J. et al. SAR11 marine bacteria require exogenous reduced sulphur for SAR11 chemoheterotrophic marine bacteria. Nat. Commun. 5:4346 doi: 10.1038/ growth. Nature 452, 741–744 (2008). ncomms5346 (2014). NATURE COMMUNICATIONS | 5:4346 | DOI: 10.1038/ncomms5346 | www.nature.com/naturecommunications 7 & 2014 Macmillan Publishers Limited. All rights reserved. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nature Communications Springer Journals

Methane production by phosphate-starved SAR11 chemoheterotrophic marine bacteria

Download PDF
7 pages

Loading next page...
 
/lp/springer-journals/methane-production-by-phosphate-starved-sar11-chemoheterotrophic-3f0WoioL3L

References (53)

Publisher
Springer Journals
Copyright
Copyright © 2014 by Nature Publishing Group, a division of Macmillan Publishers Limited. All Rights Reserved.
Subject
Science, Humanities and Social Sciences, multidisciplinary; Science, Humanities and Social Sciences, multidisciplinary; Science, multidisciplinary
eISSN
2041-1723
DOI
10.1038/ncomms5346
Publisher site
See Article on Publisher Site

Abstract

ARTICLE Received 9 Feb 2014 | Accepted 9 Jun 2014 | Published 7 Jul 2014 DOI: 10.1038/ncomms5346 Methane production by phosphate-starved SAR11 chemoheterotrophic marine bacteria 1, ,w 2, 1 1 Paul Carini * , Angelicque E. White *, Emily O. Campbell & Stephen J. Giovannoni The oxygenated surface waters of the world’s oceans are supersaturated with methane relative to the atmosphere, a phenomenon termed the ‘marine methane paradox’. The production of methylphosphonic acid (MPn) by marine archaea related to Nitrosopumilus maritimus and subsequent decomposition of MPn by phosphate-starved bacterioplankton may partially explain the excess methane in surface waters. Here we show that Pelagibacterales sp. strain HTCC7211, an isolate of the SAR11 clade of marine a-proteobacteria, produces methane from MPn, stoichiometric to phosphorus consumption, when starved for phosphate. Gene transcripts encoding phosphonate transport and hydrolysis proteins are upregulated under phosphate limitation, suggesting a genetic basis for the methanogenic phenotype. Strain HTCC7211 can also use 2-aminoethylphosphonate and assorted phosphate esters for phosphorus nutrition. Despite strain-specific differences in phosphorus utilization, these findings identify Pelagibacterales bacteria as a source of biogenic methane and further implicate phosphate starvation of chemoheterotrophic bacteria in the long-observed methane supersaturation in oxygenated waters. 1 2 Department of Microbiology, Oregon State University, Corvallis, Oregon 97331, USA. College of Earth, Ocean, and Atmospheric Sciences, Oregon State University, Corvallis, Oregon 97331, USA. * These authors contributed equally to this work. w Present address: University of Maryland Center for Environmental Sciences, Horn Point Laboratory, Cambridge, MD 21613, USA. Correspondence and requests for materials should be addressed to S.J.G. (email: [email protected]). NATURE COMMUNICATIONS | 5:4346 | DOI: 10.1038/ncomms5346 | www.nature.com/naturecommunications 1 & 2014 Macmillan Publishers Limited. All rights reserved. ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5346 he simplest of all hydrocarbons, methane (CH ), is a potent observations of CH maxima below the surface mixed layer and 4 4 greenhouse gas that has 420-fold more warming potential the lack of diurnal variation of these CH levels point to Tthan carbon dioxide, making an understanding of its heterotrophic organisms as probable contributors to CH sources and sinks important for global climate models (reviewed production—provided a steady source of MPn is available and in Valentine ). Canonical microbial methanogenesis pathways that environmental conditions favour the utilization of dissolved operate under strictly anaerobic conditions , such that organic phosphorus (DOP). Consistent with the hypothesis that methanogenesis is not favoured in marine surface waters where P-starved heterotrophic bacteria produce CH , members of oxygen is abundant. Historically, the only known sites of classical marine Vibrionales and Rhodobacterales expressed C–P lyase methanogenesis in the surface ocean were the anaerobic genes in P-limited mesocosm studies that, when amended with microenvironments of faecal matter and the guts of certain fish MPn, also generated CH (ref. 17). 3,4 or plankton . Yet, CH profiles in the open ocean show The SAR11 clade of oligotrophic a-proteobacteria (Pelagibac- subsurface maxima associated with the pycnocline (density, terales) are the numerically dominant chemoheterotrophic cells in s ¼ 25.2–25.9; 50–200 m) where CH is generally marine euphotic zones worldwide . Environmental and t 4 5,6 supersaturated in relationship to atmospheric equilibrium . laboratory studies have elucidated the forms of dissolved How this methane comes to be produced in oxygenic water organic matter Pelagibacterales bacteria use, including reduced masses is the basis of a phenomenon coined the ‘oceanic methane sulphur compounds, amino acids, one-carbon compounds, 7 29 30 paradox’ , which is particularly intriguing given that the isotopic organic acids (reviewed in Tripp ) and vitamin precursors . composition of subsurface CH maxima is inconsistent with that Despite a high degree of genomic conservation between 8 31 of anaerobic microenvironments . Pelagibacterales genomes , strategies for acquiring P appear to An alternate hypothesis has been proposed: CH is produced vary between Pelagibacterales isolates. For example, compared when marine microorganisms use methylphosphonic acid (MPn) with the northeast Pacific Ocean isolate, ‘Candidatus Pelagibacter as a resource for phosphorus (P) . Several lines of evidence ubique’ strain HTCC1062 (str. HTCC1062, herein), the genome support this ‘aerobic methane production’ hypothesis: (i) the C–P of Pelagibacterales sp. strain HTCC7211 (str. HTCC7211), lyase enzyme complex catabolizes MPn to CH , releasing isolated from the North Atlantic Ocean in the Sargasso Sea , 10,11 phosphate (P ) that can be used for growth (reviewed in encodes for additional proteins involved in organophosphorus 12 20 White and Metcalf ); (ii) C–P lyase gene sequences are present transport (PhnCDEE ) and phosphonate utilization 13–17 (sometimes abundant) in marine environments ; (iii) natural (PhnGHIJKLNM), suggesting that the two organisms have 9,17 seawater samples incubated with MPn release CH ; (iv) evolved different adaptive strategies for P acquisition related to biosynthesis of MPn by Nitrosopumilus maritimus SCM1, a niche partitioning . member of the ubiquitous and abundant Marine Group I (MGI) Here, building on these genomic insights we test the hypothesis archaea was recently demonstrated ; and (v) ca. 0.6% of that Pelagibacterales chemoheterotrophic bacteria encoding the microbes in marine surface communities contain genes C–P lyase complex produce CH from MPn when P starved. We 4 i encoding the MPn synthase . Taken together, these findings show that str. HTCC7211 produces CH from MPn in suggest the observed CH supersaturation can be explained by the proportions stoichiometric to P demand. Moreover, we present natural synthesis of MPn and subsequent hydrolysis for purposes rates of CH production under ideal growth conditions, together of P nutrition by marine plankton. with the genetic response to P starvation in the Pelagibacterales. For this process to act as a substantial source of CH , there These data suggest Pelagibacterales are one of the leading must be an ecophysiological niche for MPn utilization in the contributors to CH dynamics in P -deficient oligotrophic gyres. 4 i oxygenic surface ocean. Isolated cultures of Escherichia coli, 10,12 Pseudomonas spp. and other heterotrophic bacteria , as well as marine autotrophs such as Trichodesmium , have been shown to Results use MPn as a sole P source. In these organisms, MPn degradation The capacity of Pelagibacterales strains HTCC1062 and proceeds via the C–P lyase pathway and is induced by HTCC7211 to grow on multiple alternate P sources, including 12,20–23 P starvation . Building on these findings, studies phosphite, phosphonates and phosphate esters, was tested. Strain 9 17 by Karl et al. and Martınez et al. showed that when surface HTCC1062 grew on P and did not use added DOP (Fig. 1). In seawater is amended with glucose and a nitrogen source contrast, str. HTCC7211 used a broad range of alternate (to stimulate P -limiting conditions), MPn hydrolysis and CH compounds for P nutrition, including P esters (phosphoserine, i 4 production proceeds at rates that scale inversely with the glucose 6-phosphate and ribose 5-phosphate), phosphonates addition of exogenous P . The authors implicate a combination (2-aminoethylphosphonic acid and MPn) and reduced inorganic of heterotrophic marine bacteria, together with natural P (phosphite) (Fig. 1). In all cases, diauxic growth was observed assemblages of Trichodesmium spp., as drivers of this when str. HTCC7211 cells were grown with an alternate P source 9,17 metabolism . These findings are consistent with studies of (Fig. 2); this growth pattern is presumably related to a transition 23 24 gene expression and methane production in Trichodesmium from growth on P to growth on organic P resources as has been 33,34 erythraeum IMS101 that show induction of C–P lyase genes described for other organisms . There was no significant under P starvation and subsequent methane production from difference in negative control (no P added) cell yields and the cell MPn. yields attained when glucose 6-phosphate, ribose 5-phosphate, Although these experiments provide definitive evidence for MPn, 2-AEP and phosphoserine were supplied to str. HTCC1062 potential MPn utilization in the surface ocean, they do not (Fig. 1). We presume that this is an indication that abiotic P necessarily address the CH supersaturation observed at the release from these compounds is negligible. The increased str. pycnocline, nor the organism(s) responsible for production of HTCC1062 cell yields when grown with phosphite as a sole excess CH in these deep layers of the euphotic zone. It seems source of P could result from (i) trace amounts of phosphate; (ii) unlikely that the activity of large buoyant genera such as abiotic conversion (oxidation) of phosphite to phosphate; or (iii) Trichodesmium can wholly explain the subsurface CH features the ability of str. HTCC1062 to enzymatically oxidize phosphite given that Trichodesmium are largely observed within or at the to phosphate via an unknown mechanism. At this time we do not 25,26 base of surface mixed layer and that CH supersaturation has have evidence that supports one of these possibilities over the also been observed outside of the subtropical gyres . Alternately, other. 2 NATURE COMMUNICATIONS | 5:4346 | DOI: 10.1038/ncomms5346 | www.nature.com/naturecommunications & 2014 Macmillan Publishers Limited. All rights reserved. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5346 ARTICLE str.HTCC1062 str. HTCC7211 No P (– control) P (+ control) MPn Phosphite 2-AEP Ribose-5P Glucose-6P P-serine 5 101520253035 Time (days) Figure 2 | Diauxic growth of Pelagibacterales sp. str. HTCC7211 on alternate phosphorus sources. Points are the mean densities of biological replicates 1.0 s.d. (n¼ 3). Each alternate phosphorus compound was Figure 1 | P-source utilization by Pelagibacterales strains HTCC1062 supplied at a final concentration of 1.0mM. P was supplied as NaH PO i 2 4 (blue) and HTCC7211 (red). Bar height is the mean of the maximum density in the positive control. Growth medium was not amended with P in the achieved for biological replicates 1.0 s.d. (n¼ 3). P was supplied as negative control. Remaining treatments had 1.0mM of P source added. NaH PO in the positive control. Growth medium was not amended with P in 2 4 i Cell inoculum was previously grown on P . the negative control. Remaining treatments had 1.0mM of P source added. The genetic basis for the P source utilization patterns observed metaproteomes , suggesting an important role in the response for each strain was investigated at a transcriptional level using to P stress. Given its genomic context adjacent to an DNA microarrays. Time-course experiments showed that organophosphate transporter (phnCDEE ) and strong P -starved str. HTCC1062 cultures rapidly induced expression upregulation under P -deplete conditions, we postulate that i i of pstSCAB, encoding the high-affinity transport system (Fig. 3). PB7211_828 may be involved in P-ester hydrolysis. Individual pst genes were upregulated 2.4- to 3.6-fold 4 h after the CH is a product of MPn hydrolysis by P -starved str. 4 i onset of P starvation and 9.1- to 27.9-fold 38 h after the onset of HTCC7211 cells. When synthetic growth medium (without P starvation (Fig. 3 and Supplementary Tables 1–3). A added P ) was amended with excess MPn (10mM), str. i i concomitant upregulation of global stress-response genes (includ- HTCC7211 produced CH (Fig. 4a) concomitantly with cell ing recA, lexA and umuD) and downregulation of ribosomal growth (Fig. 4b). CH was not produced when cells were grown protein transcripts (including rpsB, rpsP, rpsK, rpsS, rplB and with P additions (Fig. 4a) and CH production from MPn was i 4 rplC) suggested that P depletion is treated as a global stressor in repressed when cells were grown in the presence of MPn and str. HTCC1062 (Supplementary Tables 2–3). P starvation of str. P (Table 1). Mass balance confirmed this observation: i i ± ± HTCC7211 induced expression of genes encoding the organo- 1.29 0.18mMCH (mean s.d.; n¼ 3; presumably from the phosphorus ABC transporter (phnCDEE ; 7.3- to 21.8-fold) and catabolism of 1.29 0.18mM MPn) was produced over the course the C–P lyase (phnGHIJKLM; 2.1- to 7.3-fold) (Fig. 3 and of the 36-day incubation, plus 8.47 0.21mM residual DOP Supplementary Tables 4 and 5). Genes encoding str. HTCC7211’s (mean s.d.; n¼ 3) measured at the end of the experiment, high-affinity P transporter pstSCAB were also upregulated, but to account for the full complement of added MPn (10mM). In a lesser degree (3.2- to 3.7-fold) than the organophosphorus culture, cell-normalized CH production was 15.2 0.04 amol transporter. In contrast to str. HTCC1062, stress-response genes CH per cell (mean s.d.; n¼ 3) at an average rate of 0.42 amol were not differentially regulated in str. HTCC7211. CH per cell per day. For comparison, the P quota of MPn-grown Although str. HTCC7211 used the P-esters glucose 6-phos- str. HTCC7211 was 10.04 0.02 amol per cell (Table 2), phate, ribose 5-phosphate and phosphoserine for growth, the suggesting that the hydrolysis of MPn exceeded the cellular P mechanism of utilization is unclear and might involve a novel demand under these ideal growth conditions. At high P source phosphatase. Genes encoding recognized bacterial alkaline concentrations (467 nM), significant growth rate differences phosphatases, commonly associated with P-ester hydrolysis were observed between the MPn and P treatments (0.29 0.00 (phoA, phoD and phoX) , are absent from the str. HTCC7211 and 0.36 0.01 per day, respectively; t-test P-value: 0.000007, genome. We speculate that str. HTCC7211 encodes a novel n¼ 6) (Table 3). Overall growth rates were slower at lower phosphatase, with a broad substrate range, that may explain the concentrations (o67 nM) of MPn or P but did not differ ± ± observed utilization of P esters. The HD-hydrolase (‘HD’ in significantly as a function of P source (0.22 0.02 and 0.24 0.04 Fig. 3b), induced by P depletion, is a member of a large family of per day, respectively; t-test P-value: 0.35, n¼ 6) (Table 3). metal-dependent phosphohydrolases and may act on P esters to release P . Interestingly, a second gene (PB7211_828; ‘828’ in Fig. 3b), divergently transcribed downstream of phnE Discussion and annotated as a ‘hypothetical protein’, was upregulated Our data reveal a differential capacity for DOP utilization among 12- to 15-fold in P -deplete conditions. Although PB7211_828 the Pelagibacterales, with str. HTCC7211, but not str. HTCC1062, is unique to str. HTCC7211 (a so-called ORFan ), peptides having a broad enzymatic capacity for DOP utilization (Fig. 1). mapping to PB7211_828 were identified in Sargasso Sea MPn was used by str. HTCC7211 in place of P (Fig. 1) and when NATURE COMMUNICATIONS | 5:4346 | DOI: 10.1038/ncomms5346 | www.nature.com/naturecommunications 3 & 2014 Macmillan Publishers Limited. All rights reserved. –P (neg. ctr) Phosphite Glucose-5P Ribose-5P MPn 2-AEP P-serine +P (pos. ctr) 10 –1 Max. density (× 10 cells L ) –1 Density (cells L ) ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5346 Ca.P. ubique str. HTCC1062 High-affinity P transport phoB phoU pstB pstA pstC pstS phoR 1181 4 h ≥30.0 20 h 2.0 Amount upregulated (fold) 38 h b Pelagibacterales sp. str. HTCC7211 High-affinity P transport Phosphonate transport C-P lyase pstS pstC pstA pstB phoU phoB HD 828 phnE2 phnE phnD phnC phnF G H phnI phnJ phnK phnL N 764 phnM 68 h 96 h Figure 3 | Comparative genomics of select P acquisition-related gene expression in P -starved cell suspensions of Pelagibacterales strains HTCC1062 (a) and HTCC7211 (b). The time elapsed since the onset of resuspension is shown to the left of each gene row. Vertical lines between genes in b represent a lack of genomic synteny. Genes shaded grey were not significantly differentially expressed. Gene names, annotations and actual expression values are listed in Supplementary Tables 1–5. 1181, SAR11_1181; HD, H-D Hydrolase; 828, PB7211_828; G, phnG;H, phnH;N, phnN, 764, PB7211_764. doing so, CH was stoichiometrically released (Fig. 4). Consistent above. In situ CH production rates are probably a function of (i) 4 4 with these observations, genes encoding enzymes and transpor- MPn supply relative to the total bioavailable DOP pool accessible ters necessary to catabolize MPn (and other phosphonates) were to the Pelagibacterales; (ii) the relative efficiency with which MPn upregulated under P limitation (Fig. 3). MPn hydrolysis was is used when multiple substrates are simultaneously available; and strongly repressed by P (Table 1), suggesting that in situ CH (iii) the percent of the Pelagibacterales population that contain C– i 4 production via MPn degradation by Pelagibacterales bacteria is P lyase genes. We speculate that maximal CH production by likely to be confined to P -deplete ocean regimes, such as the Pelagibacterales may occur shortly after the seasonal spring Sargasso Sea. Accordingly, published metagenomic analyses of mixing event in the Sargasso Sea. This hypothesis is consistent multiple oceanic environments show that the Pelagibacterales-like with the observation that MPn producers (ammonia-oxidizing C–P lyase genes are not universally distributed, but instead are MGI archaea) are most active at 120 m, the base of the euphotic 13–15 42 more abundant in P -deplete waters . zone in the Sargasso Sea , and that Pelagibacterales stocks Although Pelagibacterales are likely to contribute meaningfully increase sharply after spring mixing . to the observed CH supersaturation in surface waters because of The laboratory results presented here help to contextualize 13–15 38 their large population size, calculating their specific contribution metagenomic and metaproteomic findings that suggested is challenging owing to the uncertainties pertaining to that Pelagibacterales bacteria actively transport and use DOP, strain heterogeneity with respect to C–P lyase gene content and including phosphonates and P-esters, in the Sargasso Sea (Fig. 1). how that may relate to the observed niche partitioning of The significance of these findings is extended to the global climate Pelagibacterales ecotypes. In the Sargasso Sea, total Pelagibacter- by showing that P -starved Pelagibacterales produce CH when i 4 8  1 ales cell abundances oscillate between 0.5 and 2.2 10 cells l MPn is provided as a sole P source (Fig. 4). Interestingly, in with maximum abundances at or above the pycnocline . addition to the abundant MGI archaea, some Pelagibacterales Spatiotemporal variation in the distribution of Pelagibacterales ecotypes may also harbour genes encoding the MPn synthase 31,40,41 18 ecotypes has been discussed in multiple studies . The degree enzyme , suggesting Pelagibacterales may also be an MPn source to which the presence of C–P lyase genes correlates with the in the ocean. An understanding of environmental MPn dynamics different Pelagibacterales ecotypes and whether the relative (including supply and consumption rates) as it pertains to DOP abundances of Pelagibacterales-like C–P lyase genes fluctuate utilization, remains a significant challenge to resolving CH fluxes seasonally is unknown. When analysing pyrosequences from as they pertain to P stress in marine environments. the Sargasso Sea, Martinez et al. showed that 82.5% of Although methane production by P -starved bacterioplankton the identifiable phnI genes (used as a proxy for the presence is a probable source of CH in P -deplete waters rich in MPn, 4 i of the C–P lyase) best hit to str. HTCC7211’s phnI gene, there may be other CH -producing biological pathways that bear suggesting that Pelagibacterales members mediate a large portion consideration. For example, in the nitrate-deplete, P -replete of the C–P lyase activity in the Sargasso Sea. Using a metric called waters of the central Arctic Ocean, an inverse correlation ‘multiplicity per cell,’ Coleman and Chisholm analysed between CH and a common resource for bacterial carbon and metagenomic data from the Sargasso Sea (collected in October sulphur, dimethylsulphoniopropionate (DMSP), was found .In of 2006) and calculated that B52% of all Pelagibacterales cells subsequent incubation experiments, when DMSP was added contained phnJ, a marker gene used to infer the presence of the to the Arctic surface waters, CH was produced and a- and 16 27 C–P lyase gene suite . Multiplying the approximate abundance g-proteobacteria increased in relative abundance . From these of C–P lyase containing Pelagibacterales cells found in results, Damm et al. hypothesized a type of ‘methylotrophic the P-deplete Sargasso Sea (52%, based on phnJ,of methanogenesis’ in which the methylated degradation products of 8  1 0.5–2.2 10 cells l ) by the maximal net cell-specific CH DMSP (methanethiol in particular) act as a precursor for CH 4 4 production rate measured here (0.42 amol CH per cell per day; production. Although this is an intriguing possibility that may Fig. 4), we estimate maximal CH production potential of ca. represent an alternate, and potentially significant, CH source in 4 4 0.01–0.05 nM per day via MPn hydrolysis. aerobic surface waters, the enzymes, pathways and organisms As MPn is part of a complex milieu of dissolved organopho- involved have not yet been identified. Further study of sphorus compounds that Pelagibacterales can use (for example, mechanisms of aerobic CH production in relationship to those illustrated in Fig. 1), the actual CH production rates are bacterial diversity and gene expression are needed to further probably much lower than the potential contribution calculated unravel oceanic CH sources. 4 NATURE COMMUNICATIONS | 5:4346 | DOI: 10.1038/ncomms5346 | www.nature.com/naturecommunications & 2014 Macmillan Publishers Limited. All rights reserved. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5346 ARTICLE Table 1 | CH production by str. HTCC7211 from MPn (10 lM) as a function of P addition*. 1.4 P added (lM) Net CH produced (lM) % Repressionw i 4 1.2 0 1.29 0 0.08 0.76 41 1.0 0.125 0.14 89 0.25 0.05 96 0.8 0.5 0.05 96 0.75 0.04 97 0.6 1 0.04 97 i 5 0.03 97 0.4 10 0.02 98 MPn 0.2 MPn, methylphosphonic acid. *Values after 36-day incubation. 0 w% Repression¼ 100 (1 (net CH produced/1.29)). Table 2 | Pelagibacterales sp. str. HTCC7211 cell yield as a function of added P. P treatment MPn treatment Added P , nM Maximum Added MPn, nM Maximum 1  1 cell yield (l )* cell yield (l )* 9 9 ± ± 98 9.41 0.36 10 67 7.31 0.23 10 9 9 ± ± i 9.8 1.21 0.14 10 6.7 1.30 0.05 10 8 8 ± ± MPn 0.98 6.78 1.53 10 0.67 7.14 0.23 10 No P ± ± i P quotaw 10.95 0.30 amol P quotaw 10.04 0.02 amol 1  1 1 P cell P cell MPn, methylphosphonic acid. *Cell yields are presented as the mean s.d. (n¼ 3). wThe P quota of str. HTCC7211 cells was determined via a dose–response curve using cultures grown on P or MPn. Triplicate cultures were created for each P dose and grown under the conditions described in ‘cultivation methods’ until P-limited stationary phase was reached. 0.12 The P quota was derived from linear regression s.d. through the maximum cell yield versus the initial dissolved P added. 0.10 0.08 Table 3 | Specific growth rates of str. HTCC7211 in AMS1 0.06 with varying levels of P or MPn. 0.04 P treatment MPn treatment CH 4 Added Specific growth rate Added Specific growth rate 0.02 2 P,nM (per day)* MPn, nM (per day)* Cell i ± ± 9800 0.35 0.01 6700 0.29 0.00 ± ± 98 0.36 0.01 67 0.29 0.01 0 ± ± 9.8 0.27 0.01 6.7 0.24 0.01 0 5 10 15 20 25 30 35 40 ± ± 0.98 0.20 0.02 0.67 0.21 0.01 Days ± ± Mean 0.29 0.07 Mean 0.26 0.04 Figure 4 | Pelagibacterales str. HTCC7211 produces methane when P MPn, methylphosphonic acid. starved. (a) Total CH concentration in sealed bottles from Pelagibacterales ± *Specific growth rates are presented as the mean s.d. (n¼ 3). sp. str. HTCC7211 cultures grown with MPn or P . Points are the mean CH concentration 1.0 s.d. (n¼ 3) (b) Cell densities from bottles of 4nM; B : 700 pM; Myo-inositol: 6mM; 4-aminobenzoate: 60 nM). P sources were Pelagibacterales sp. str. HTCC7211 cultures grown with no P , MPn or P.Cell i i added as indicated through the main text. counts for each treatment were obtained from single bottles used solely for cell counts. (c) The CH production rate (change in CH concentration per 4 4 day) and cell production rate (change in cell density per day) from MPn Cultivation details. All cultures, except those describing CH production (described below), were grown in acid-washed and autoclaved polycarbonate flasks. treatments before day 25, after which cell concentrations began to decline. Cultures were incubated at 20 C with shaking at 60 r.p.m. under a 12-h light:12-h dark cycle. Light levels during the day were held at 140–180mmol photons 2  1 m s . Cells for cell counts were stained with SYBR green I and counted with a Methods Guava Technologies flow cytometer at 48–72 h intervals as described elsewhere . Organism source. ‘Candidatus Pelagibacter ubique’ str. HTCC1062 (ref. 43) and The treatments for the alternate phosphate source utilization experiments were: no Pelagibacterales sp. str. HTCC7211 (ref. 32) were revived from 10% glycerol stocks P addition (negative control), 1.0mMP (as NaH PO ; positive control), 1.0mM i i 2 4 and propagated in artificial seawater medium for SAR11 (AMS1) , without added phosphite (as NaPHO 5H O), 1.0mM glucose 6-phosphate, 1.0mM ribose 3 2 P , but amended with pyruvate (100mM), glycine (5mM), methionine (5mM), FeCl 5-phosphate, 1.0mM MPn, 1.0mM 2-aminoethylphosphonic acid or 1.0 mM i 3 (1mM), and vitamins (B :6mM; B : 800 nM; B : 425 nM; B : 500 nM; B :4nM;B : phosphoserine. All alternate phosphorus compounds were Z98% pure. 1 3 5 6 7 9 NATURE COMMUNICATIONS | 5:4346 | DOI: 10.1038/ncomms5346 | www.nature.com/naturecommunications 5 & 2014 Macmillan Publishers Limited. All rights reserved. –1 CH production rate (nM d ) –1 10 Total CH (μM) Density (cells L × 10 ) –1 –1 9 Cell production rate (L d × 10 ) ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5346 Methane production from MPn. Pelagibacterales sp. str. HTCC7211 cells, pre- References viously grown in media with MPn as the sole P source, were used to inoculate 1. Valentine, D. L. Emerging topics in marine methane biogeochemistry. Annu. sterile, BSA-coated glass serum bottles containing growth media amended with Rev. Marine. Sci. 3, 147–171 (2011). 10mM MPn as the sole P source. The negative control consisted of str. HTCC7211 2. Reeburgh, W. S. Oceanic methane biogeochemistry. Chem. Rev. 107, 486–513 cells, previously grown in media with P as the sole P source, as inoculum for i (2007). sterile, BSA-coated serum bottles containing growth media amended with 10mMP 3. Tilbrook, B. D. & Karl, D. M. Methane sources, distributions and sinks from as the sole P source. Serum bottles (60 ml) were filled with 55 ml culture, leaving a California coastal waters to the oligotrophic North Pacific gyre. Mar. Chem. 49, 5-ml headspace, capped with Viton septa, crimped with aluminum seals and 51–64 (1995). incubated horizontally in the conditions described in ‘Cultivation details.’ Cell 4. Sasakawa, M. et al. Carbon isotopic characterization for the origin of excess counts for each treatment were obtained from single bottles used solely for methane in subsurface seawater. J. Geophys. Res. 113, C03012 (2008). cell counts. 5. Burke, R. J., Reid, D. F., Brooks, J. M. & Lavoie, D. M. Upper water column CH in the headspace was measured by gas chromatography using a Shimadzu methane geochemistry in the Eastern Tropical North Pacific. Limnol. Oceanogr. GC-8A gas chromatograph equipped with a column packed with Porapak N 28, 19–32 (1983). (80/100 mesh size) fitted with a flame ionization detector. In brief, 100ml samples 6. Holmes, M. E., Sansone, F. J., Rust, T. M. & Popp, B. N. Methane production, of headspace from each bottle were injected into the GC at a flow rate of 1 consumption, and air-sea exchange in the open ocean: An Evaluation based on 25 ml min using ultrapure helium (He) as the carrier gas. Peaks were integrated carbon isotopic ratios. Global Biogeochem. Cycles 14, 1–10 (2000). using PeakSimple chromatography software (http://www.srigc.com/). 7. Kiene, R. P. in Microbial Production and Consumption of Greenhouse Gases Quantification of CH was accomplished by calibrating peak areas to the flame Methane, Nitrogen Oxides, and Halomethanes 111–146 (ASM, 1991). ionization detector response to a three-point calibration conducted at the beginning and end of the experiment using 100ml injections of 100, 1,000 and 8. Sansone, F. J., Graham, A. W. & Berelson, W. M. Methane along the western 10,000 p.p.m. CH standards. The 1,000 and 10,000 p.p.m. standards were prepared Mexican margin. Limnol. Oceanogr. 49, 2242–2255 (2004). from a 1% mixed gas standard containing CH , ethane and ethylene. Replicates of 4 9. Karl, D. M. et al. Aerobic production of methane in the sea. Nat. Geosci. 1, the 100 p.p.m. standard were analysed alongside samples on each sampling day. 473–478 (2008). Over the course of the experiment, the peak area for the 100 p.p.m. standard varied 10. Daughton, C. G., Cook, A. M. & Alexander, M. Biodegradation of phosphonate by o5% (coefficient of variation). toxicants yields methane or ethane on cleavage of the C-P bond. FEMS Headspace equilibration models at 20 C using a Bunsen coefficient of 0.0345 Microbiol. Lett. 5, 91–93 (1979). for a 60 ml total volume (55 ml liquid, 5 ml headspace) suggest the fraction of the 11. Kamat, S. S., Williams, H. J. & Raushel, F. M. Intermediates in the methane in the gas phase to be 74.5%. The fractionation of CH between headspace transformation of phosphonates to phosphate by bacteria. Nature 480, 570–573 and aqueous phase was verified by injecting commercial CH standards (1.0 ml of (2011). either 100 p.p.m. or 10,000 p.p.m. CH equivalent to 4.08 nmol CH , 40.8 nmol 4 4 12. White, A. K. & Metcalf, W. W. Microbial metabolism of reduced phosphorus CH and 408 nmol CH , respectively) into sealed serum bottles containing 4 4 compounds. Annu. Rev. Microbiol. 61, 379–400 (2007). uninoculated growth medium. Of the added CH ,72 6% in these standards was 13. Martinez, A., Tyson, G. W. & Delong, E. F. Widespread known and novel measured in the headspace at 20 C. A value of 74% was used to adjust headspace phosphonate utilization pathways in marine bacteria revealed by functional CH measurements to total CH . 4 4 screening and metagenomic analyses. Environ. Microbiol. 12, 222–238 (2010). 14. Coleman, M. L. & Chisholm, S. W. Ecosystem-specific selection pressures revealed through comparative population genomics. Proc. Natl Acad. Sci. 107, Dissolved phosphorus measurements. Samples for determination of total 18634–18639 (2010). dissolved P levels were obtained at the end of CH bioassays (see ‘Methane pro- 15. Martiny, A. C., Huang, Y. & Li, W. in Handbook of Molecular Microbial Ecology duction from MPn’). Cells were removed from spent growth medium by filtration II: Metagenomics in Different Habitats 269–276 (John Wiley & Sons, Inc., through a 0.02-mm Acrodisc filter. The dissolved inorganic P (DIP) concentration of the filtrate was determined colorimetrically as described by Murphy and Riley 2011). using a Cary ultraviolet–visible spectrophotometer. The total dissolved P 16. Villarreal-Chiu, J. F., Quinn, J. P. & Mcgrath, J. W. The genes and enzymes of concentration of the filtrate was measured using a modification of the phosphonate metabolism by bacteria, and their distribution in the marine Valderrama method . Briefly, samples were oxidized at 120 C for 60 min in environment. Front. Microbiol. 3, 19–19 (2012). Teflon digestion bombs, allowed to cool and run as DIP, as described above. 17. Martınez, A., Ventouras, L. A., Wilson, S. T., Karl, D. M. & DeLong, E. F. The oxidizing agent used was a solution of potassium persulphate, sodium Metatranscriptomic and functional metagenomic analysis of hydroxide and boric acid (as per ref. 46) added in a 1:10 oxidant to sample methylphosphonate utilization by marine bacteria. Front. Microbiol. 4, 340 ratio. DOP concentrations were calculated as the difference between total dissolved (2013). P and DIP. 18. Metcalf, W. W. et al. Synthesis of methylphosphonic acid by marine microbes: a source for methane in the aerobic ocean. Science (New York, NY) 337, 1104–1107 (2012). Microarray growth and sampling conditions. Biological replicates (n¼ 3) of str. 19. White, A. E., Karl, D. M., Bjo¨rkman, K., Beversdorf, L. J. & Letelier, R. M. HTCC1062 or str. HTCC7211 were grown in AMS1 (ref.39) amended with Production of organic matter by Trichodesmium IMS101 as a function of pyruvate (1 mM), glycine (50mM), methionine (50mM), FeCl (1mM), P (10mM) 3 i phosphorus source. Limnol. Oceanogr. 55, 1755–1767 (2010). and vitamins (as described above). In mid- to late-logarithmic growth (ca. 0.7- 8  1 20. Metcalf, W. W. & Wanner, B. L. Involvement of the Escherichia coli phn (psiD) 1.0 10 cells ml ), cells were harvested by centrifugation (17,664g for 1.0 h at gene cluster in assimilation of phosphorus in the form of phosphonates, 20 C). Cell pellets were evenly split and washed twice with growth media that was phosphite, Pi esters, and Pi. J. Bacteriol. 173, 587–600 (1991). either amended with 100mMP (replete conditions) or not amended with P i i 21. Wanner, B. L. & Metcalf, W. W. Molecular genetic studies of a 10.9-kb operon (deplete conditions). After washing, pellets were re-suspended in P -deplete or in Escherichia coli for phosphonate uptake and biodegradation. FEMS P -replete growth media. Samples were collected from cell suspensions at t¼ 0, 4, 20 and 38 h after resuspension for str. HTCC1062, and t¼ 0, 20, 38, 68 and 96 h Microbiol. Lett. 100, 133–139 (1992). after resuspension for str. HTCC7211. Sampling consisted of centrifugation 22. Metcalf, W. W. & Wanner, B. L. Mutational analysis of an Escherichia coli (48,298g for 1.0 h at 4 C) followed by resuspension of cell pellets in ‘RNAprotect fourteen-gene operon for phosphonate degradation, using TnphoA’ elements. Bacteria’ reagent (Qiagen, Valencia, CA). RNA was extracted using an RNeasy J. Bacteriol. 175, 3430–3442 (1993). Mini kit (Qiagen) and amplified using the MessageAmp-II Bacteria RNA 23. Dyhrman, S. et al. Phosphonate utilization by the globally important marine amplification kit (Ambion, Carlsbad, CA) using the manufacturer’s instructions. diazotroph Trichodesmium. Nature 439, 68–71 (2006). Amplified and labelled RNA was hybridized to custom ‘Candidatus Pelagibacter 24. Beversdorf, L. J., White, A. E., Bjo¨rkman, K. M., Letelier, R. M. & Karl, D. M. ubique’ Affymetrix GeneChip arrays (Pubiquea520471f) (Affymetrix, Santa Clara, Phosphonate metabolism by Trichodesmium IMS101 and the production of CA) that contained probes for Pelagibacterales strains HTCC1002, HTCC1062 and greenhouse gases. Limnol. Oceanogr. 55, 1768–1778 (2010). HTCC7211 as described previously . Briefly, Affymetrix GeneChip arrays were 25. Davis, C. S. & McGillicuddy, D. J. Transatlantic abundance of the N2-fixing hybridized and scanned using an Affymetrix GeneChip fluidics station, colonial cyanobacterium Trichodesmium. Science (New York, NY) 312, 1517– hybridization oven and scanner. Background and raw expression values were 1520 (2006). normalized between chips using the robust microarray algorithm using the 26. Letelier, R. M. & Karl, D. M. Role of Trichodesmium spp. in the productivity of Affymetrix Expression Console software. A Bayesian statistical analysis was 48 the subtropical North Pacific Ocean. Mar. Ecol. Prog. Ser. 133, 263–273 (1996). conducted using Cyber-T . The estimate of variance was calculated in 27. Damm, E. et al. Methane production in aerobic oligotrophic surface water in Cyber-T by using window sizes of 101 and a confidence value of 10. A t-test was the central Arctic Ocean. Biogeosciences 7, 1099–1108 (2010). performed on log-transformed expression values by using the Bayesian variance 28. Morris, R. et al. SAR11 clade dominates ocean surface bacterioplankton estimate. The programme QVALUE, was used to obtain a q-value, which accounts for multiple t-tests performed . A gene was defined as differentially expressed communities. Nature 420, 806–810 (2002). 29. Tripp, H. J. The unique metabolism of SAR11 aquatic bacteria. J. Microbiol. 51, if both the q-value was r0.05 and the gene was differentially regulated by Z2.0-fold. 147–153 (2013). 6 NATURE COMMUNICATIONS | 5:4346 | DOI: 10.1038/ncomms5346 | www.nature.com/naturecommunications & 2014 Macmillan Publishers Limited. All rights reserved. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5346 ARTICLE 30. Carini, P. et al. Discovery of a SAR11 growth requirement for thiamin’s 45. Murphy, J. & Riley, J. P. A modified single solution method for the pyrimidine precursor and its distribution in the Sargasso Sea. ISME J. determination of phosphate in natural waters. Anal. Chim. Acta 27, 31–36 (1962). doi:10.1038/ismej.2014.61. 46. Valderrama, J. C. The simultaneous analysis of total nitrogen and total 31. Grote, J. et al. Streamlining and core genome conservation among highly phosphorus in natural waters. Mar. Chem. 10, 109–122 (1981). divergent members of the SAR11 clade. MBio 3, e00252–12 (2012). 47. Schwalbach, M. S., Tripp, H. J., Steindler, L., Smith, D. P. & Giovannoni, S. J. 32. Stingl, U., Tripp, H. J. & Giovannoni, S. J. Improvements of high-throughput The presence of the glycolysis operon in SAR11 genomes is positively correlated culturing yielded novel SAR11 strains and other abundant marine bacteria from with ocean productivity. Environ. Microbiol. 12, 490–500 (2010). the Oregon coast and the Bermuda Atlantic Time Series study site. ISME J. 1, 48. Baldi, P. & Long, A. D. A Bayesian framework for the analysis of microarray 361–371 (2007). expression data: regularized t -test and statistical inferences of gene changes. 33. Daughton, C. G., Cook, A. M. & Alexander, M. Phosphate and soil binding: Bioinformatics 17, 509–519 (2001). factors limiting bacterial degradation of ionic phosphorus-containing pesticide 49. Storey, J. D. & Tibshirani, R. Statistical significance for genomewide studies. metabolites. Appl. Environ. Microbiol. 37, 605–609 (1979). Proc. Natl Acad. Sci. 100, 9440–9445 (2003). 34. Wackett, L. P., Wanner, B. L., Venditti, C. P. & Walsh, C. T. Involvement of the phosphate regulon and the psiD locus in carbon-phosphorus lyase activity of Escherichia coli K-12. J. Bacteriol. 169, 1753–1756 (1987). Acknowledgements 35. Luo, H., Benner, R., Long, R. A. & Hu, J. Subcellular localization of We thank Katie Watkins-Brandt for P measurements and Kiriann L. Carini for con- ceptual thumbnail. Microarrays were hybridized at the Oregon State University Center marine bacterial alkaline phosphatases. Proc. Natl Acad. Sci. 106, 21219–21223 for Genome Research and Biocomputing Core Laboratory. This research was funded by (2009). 36. Aravind, L. & Koonin, E. V. The HD domain defines a new superfamily of the Gordon and Betty Moore Foundation through Grant GBMF607.01 to Stephen Giovannoni and National Science Foundation grants OCE-0802004 and OCE-0962362. metal-dependent phosphohydrolases. Trends Biochem. Sci. 23, 469–472 (1998). 37. Fischer, D. & Eisenberg, D. Finding families for genomic ORFans. Bioinformatics 15, 759–762 (1999). Author contributions 38. Sowell, S. M. et al. Transport functions dominate the SAR11 metaproteome at P.C., A.W. and S.J.G. designed experiments and prepared manuscript. P.C., A.W. and low-nutrient extremes in the Sargasso Sea. ISME J. 3, 93–105 (2009). E.O.C. conducted experiments. 39. Carini, P., Steindler, L., Beszteri, S. & Giovannoni, S. J. Nutrient requirements for growth of the extreme oligotroph ‘Candidatus Pelagibacter ubique’ Additional information HTCC1062 on a defined medium. ISME J. 7, 592–602 (2013). Accession codes: The raw microarray data files are deposited in the NCBI Gene 40. Carlson, C. A. et al. Seasonal dynamics of SAR11 populations in the euphotic Expression Omnibus (GEO) database with accession codes GSM1318808 to and mesopelagic zones of the northwestern Sargasso Sea. ISME J. 3, 283–295 GSM1318861. (2009). 41. Vergin, K. L. et al. High-resolution SAR11 ecotype dynamics at the Bermuda Supplementary Information accompanies this paper at http://www.nature.com/ Atlantic Time-series Study site by phylogenetic placement of pyrosequences. naturecommunications ISME J. 7, 1322–1332 (2013). Competing financial interests: The authors declare no competing financial interests. 42. Newell, S. E., Fawcett, S. E. & Ward, B. B. Depth distribution of ammonia oxidation rates and ammonia-oxidizer community composition in the Sargasso Reprints and permission information is available online at http://npg.nature.com/ Sea. Limnol. Oceanogr 58, 1491–1500 (2013). reprintsandpermissions/ 43. Rappe´, M. S., Connon, S. A., Vergin, K. L. & Giovannoni, S. J. Cultivation of the ubiquitous SAR11 marine bacterioplankton clade. Nature 418, 630–633 (2002). How to cite this article: Carini, P. et al. Methane production by phosphate-starved 44. Tripp, H. J. et al. SAR11 marine bacteria require exogenous reduced sulphur for SAR11 chemoheterotrophic marine bacteria. Nat. Commun. 5:4346 doi: 10.1038/ growth. Nature 452, 741–744 (2008). ncomms5346 (2014). NATURE COMMUNICATIONS | 5:4346 | DOI: 10.1038/ncomms5346 | www.nature.com/naturecommunications 7 & 2014 Macmillan Publishers Limited. All rights reserved.

Journal

Nature Communications – Springer Journals

Published: Jul 7, 2014

There are no references for this article.