Hydroxylamine As an Intermediate in Ammonia Oxidation by Globally Abundant Marine Archaea

Hydroxylamine as an intermediate in ammonia oxidation by globally abundant marine archaea

Neeraja Vajralaa,1, Willm Martens-Habbenab,1, Luis A. Sayavedra-Sotoa, Andrew Schauerc, Peter J. Bottomleyd, David A. Stahlb, and Daniel J. Arpa,2

Departments of aBotany and Plant Pathology and dMicrobiology, Oregon State University, Corvallis, OR 97331; and Departments of bCivil and Environmental Engineering and cEarth and Space Science, University of Washington, Seattle, WA 98195

Edited by Edward F. DeLong, Massachusetts Institute of Technology, Cambridge, MA, and approved December 7, 2012 (received for review August 17, 2012)

The ammonia-oxidizing archaea have recently been recognized as or cytochromes c554 and cM552 critical for energy conversion in a significant component of many microbial communities in the AOB (8–15). Thus, either the product of NH3 oxidation is not biosphere. Although the overall stoichiometry of archaeal chemo- NH2OH or, alternatively, these phylogenetically deeply branch- − autotrophic growth via ammonia (NH3) oxidation to nitrite (NO2 ) ing thaumarchaea use a novel biochemistry for NH2OH oxida- is superficially similar to the ammonia-oxidizing bacteria, genome tion and electron transfer (8). sequence analyses point to a completely unique biochemistry. The In an attempt to gain further insights into the biochemistry and only genomic signature linking the bacterial and archaeal bioche- physiology of these unique archaeal nitrifiers, we here investigated Nitrosopumilus maritimus mistries of NH3 oxidation is a highly divergent homolog of the the role of NH2OH in metabolism. ammonia monooxygenase (AMO). Although the presumptive These studies were complicated by the extremely oligotrophic product of the putative AMO is hydroxylamine (NH2OH), the ab- character of this organism contributing to very low cell densities sence of genes encoding a recognizable ammonia-oxidizing bacte- in culture (16). To overcome the challenge of working with low ria-like hydroxylamine oxidoreductase complex necessitates either cell density cultures of N. maritimus, we established a method to a novel enzyme for the oxidation of NH2OH or an initial oxidation concentrate cells on nylon membrane filters such that the cells − product other than NH2OH. We now show through combined remained competent for NH3-dependent NO2 formation and

physiological and stable isotope tracer analyses that NH2OH is oxygen (O2) uptake. This method enabled us to carry out relatively SCIENCES

both produced and consumed during the oxidation of NH3 to short duration physiological studies and stable isotope tracer ENVIRONMENTAL − NO2 by Nitrosopumilus maritimus, that consumption is coupled experiments directed at determining if N. maritimus can oxidize − to energy conversion, and that NH2OH is the most probable prod- exogenous NH2OH to NO2 while consuming O2 and producing uct of the archaeal AMO homolog. Thus, despite their deep phy- ATP, and if NH2OH is an intermediate in NH3 oxidation pathway logenetic divergence, initial oxidation of NH3 by bacteria and of N. maritimus. archaea appears mechanistically similar. They however diverge bio-

chemically at the point of oxidation of NH2OH, the archaea possibly Results − catalyzing NH2OH oxidation using a novel enzyme complex. N. maritimus Oxidizes NH2OH to NO2 . The apparent lack of genes encoding a recognizable AOB-like HAO complex in the genome − N. maritimus icrobial oxidation of ammonia (NH ) to nitrite (NO ), the of has generated considerable interest in the 3 2 fi Mfirst step in nitrification, plays a central role in the global pathway of NH3 oxidation in AOA (8). Because puri cation and cycling of nitrogen. Recent studies have established that marine direct biochemical characterization of AMO enzymes has thus far been unsuccessful, we aimed to determine if N. maritimus and terrestrial representatives of an abundant group of archaea, − could convert NH2OH to NO2 . To determine if N. maritimus now classified as Thaumarchaeota, are autotrophic NH3 oxidizers − – can oxidize NH2OH to NO2 , it was necessary to expose the (1 5). Despite increasing evidence that ammonia-oxidizing N. maritimus + archaea (AOA) generally outnumber ammonia-oxidizing bacte- cells to NH2OH in the absence of the NH4 used to grow the culture. Hence, a technique was developed to con- ria (AOB), and likely nitrify in most natural environments, very μ fi little is known about their physiology or supporting biochemistry centrate the cells onto 0.2- m pore size nylon membrane lters (6, 7). Genome sequence analyses have pointed to a unique using a vacuum manifold system. The manifold system allowed simultaneous filtration of several liters of culture. In addition to pathway for NH3 oxidation, likely using copper as a major redox + active metal, and coupled to a variant of the hydroxypropionate/ concentrating the cells and separating them from residual NH4 in the medium, the nylon membranes served as a solid support hydroxybutyrate cycle (8). However, the only genome sequence for convenient handling of concentrated cells in subsequent in- feature that associates the archaeal pathway for NH oxidation 3 cubation assays. The membrane-associated N. maritimus cells ac- with that of the better characterized AOB is a divergent variant tively oxidized NH for several hours. of the ammonia monooxygenase (AMO), which may or may not 3 To investigate whether NH2OH can be oxidized by N. maritimus, be a functional equivalent of the bacterial AMO. Thus, the − a series of short-term NO accumulation assays was conducted, supporting biochemistry of a biogeochemically significant group 2 in which membrane-associated N. maritimus cells were exposed of microorganisms remains unresolved (8, 9). to various concentrations of NH OH. Irrespective of the initial Among the AOB, as represented by the model organism 2 Nitrosomonas europaea,NH3 is first oxidized to hydroxylamine (NH OH) by AMO, an enzyme composed of three subunits 2 Author contributions: N.V., W.M.-H., L.A.S.-S., P.J.B., D.A.S., and D.J.A. designed research; encoded by amoC, amoA, and amoB genes (7). NH2OH is sub- − N.V., W.M.-H., L.A.S.-S., and D.A.S. performed research; A.S. contributed new reagents/ sequently oxidized to NO2 by the hydroxylamine oxidoreductase analytic tools; N.V., W.M.-H., L.A.S.-S., A.S., P.J.B., D.A.S., and D.J.A. analyzed data; and (HAO) (7), a heme-rich enzyme encoded by the hao gene (7). Of N.V., W.M.-H., L.A.S.-S., and D.A.S. wrote the paper. fl the four electrons released from the oxidation of NH2OH to The authors declare no con ict of interest. − NO2 , two are transferred to the terminal oxidase for respiratory This article is a PNAS Direct Submission. purposes and two are transferred to AMO for further oxidation 1N.V. and W.M.-H. contributed equally to this work. of NH3 (7). Although all available genome sequences for the 2To whom correspondence should be addressed. E-mail: [email protected]. amoB amoC AOA contain homologs of the bacterial AMO ( , , This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. and amoA), there are no obvious homologs of AOB-like HAO, 1073/pnas.1214272110/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1214272110 PNAS Early Edition | 1of6 Downloaded by guest on September 25, 2021 concentration of NH2OH (50–200 μM), N. maritimus oxidized − ∼50 μMNH2OH to NO2 over 20 h (Fig. 1). During the first 30 min of exposure to 200 μMNH2OH, the rate of NH2OH- − dependent NO2 production was equal to the rate of NH3-de- − pendent NO2 production [0.19 ± 0.01 μmol/(min × mg protein)]. With longer incubation times, the rate of NH2OH oxidation decreased gradually and finally stopped after accumulation of − ∼50 μMofNO2 (Fig. 1). Addition of 1 mM NH2OH led to a − complete inhibition of the NO2 formation (Fig. 1), reflecting a toxicity at higher concentrations also observed for AOB (17). Attempts to grow N. maritimus on NH2OH as sole energy source failed as previously observed in N. europaea (18). Acetylene (C2H2) (19) and allylthiourea (ATU) (20) are specific inhibitors of NH3 oxidation by AMO in AOB and AOA (19–21). In the present work, the ability of N. maritimus to oxidize NH3 or NH2OH in presence of either C2H2 or ATU was analyzed − by monitoring NO2 accumulation. C2H2 (0.1%) was added to the headspace of sealed bottles containing membrane-attached N. maritimus cells in synthetic crenarchaeota medium (SCM) con- + taining either NH4 or NH2OH. C2H2 completely inhibited NH3- − dependent NO2 accumulation but had no effect on NH2OH- − dependent NO2 production (Fig. S1 A and B)aspreviously observed in N. europaea (19). Similarly, the addition of ATU also − inhibited NH3-dependent NO2 production but did not inhibit − NH2OH-dependent NO2 production (Fig. S1 A and B)(20). These results clearly indicate that N. maritimus can oxidize Fig. 2. Stoichiometry of NH2OH oxidation in N. maritimus.O2 uptake (solid − NH2OH and its oxidation is independent of the AMO activity squares), NH2OH consumed (open circles), NO2 produced (gray triangles); (Fig. S1 A and B). Data are means of triplicates, with variation of less than 10%. Incubations were carried out for 8 min while the conditions were at optimum (i.e., the NH2OH Oxidation Is Coupled to O2 Uptake in N. maritimus. Further unavoidable degradation of NH2OH did not interfere with the stoichiome- experiments were conducted to determine if oxidation of NH2OH try). The experiment was repeated at least three times and produced similar − results. Error bars represent SEM. to NO2 by N. maritimus is accompanied by consumption of a stoichiometric amount of O2. To test this, NH2OH consumption, − O2 uptake, and NO2 accumulation were independently deter- 1NH3:1.5 O2 previously reported for both the bacterial and ar- mined (Fig. 2). The rate of NH2OH-dependent O2 consumption − by N. maritimus remained constant for 8 min at a rate of 0.20 ± chaeal oxidation of NH3 to NO2 (16, 23). Substrate depleted N. maritimus cells consumed only 0.003 ± 0.001 μmol O2 per 0.01 μmol O2/(min × mg protein) and the ratio of O2 consumed to − minute per milligram protein. This endogenous rate of O2 uptake NO2 produced was 1 ± 0.04 (Fig. 2). This stoichiometry is similar by N. maritimus was 50–100 times lower compared with the O2 to that observed in AOB (22). By comparison, the rate of O2 + uptake rate observed with NH3 or NH2OH and consequently consumption in the presence of NH4 (200 μM) was 0.28 ± 0.03 provides a good measure of the O2 uptake rates observed during μmol O2/(min × mg protein) and reflects the stoichiometry of NH3 or NH2OH oxidation. − To evaluate the possibility that NO2 was produced from NH2OH before inhibition by C2H2 or ATU was maximal, NH3- dependent and NH2OH-dependent O2 consumption were studied in the presence of either C2H2 or ATU. Membrane-immobilized N. maritimus cells were treated with C2H2 for 2 h before placing them in the O2 electrode chamber. C2H2-exposed cells com- menced respiring immediately after the addition of NH2OH (Fig. S2A). Similarly, ATU-inhibited N. maritimus cells immediately resumed O2 consumption upon addition of NH2OH (Fig. S2B). In agreement with previous observations (2), higher concentrations of ATU (2.5 mM) were required in our studies as well to com- − pletely stop NH3-dependent NO2 accumulation and O2 uptake in N. maritimus, whereas as little as 10 μM ATU concentrations − inhibited NH3-dependent NO2 accumulation and O2 uptake in N. europaea (20). The fact that C2H2 or ATU did not inhibit NH2OH-dependent O2 uptake clearly indicates that NH2OH oxidation is not dependent on AMO (Fig. S2 A and B).

NH2OH Oxidation Is Coupled to ATP Production in N. maritimus. If N. maritimus − NH2OH is an intermediate of NH3 oxidation in Fig. 1. Accumulation of NO2 by N. maritimus. Determinations in the analogous to the AOB, then the oxidation of NH OH should be μ + μ 2 presence of 200 MNH4 (dashed lines, squares), 200 MNH2OH (solid lines, coupled to the respiratory electron transfer chain and ultimately diamonds), 50 μMNH2OH (solid lines, triangles), 1 mM NH2OH (solid lines, circles), and no substrate (dashed lines, crosses). Inset shows accumulation of to the reduction of O2, and to ATP synthesis. Indeed, we dem- − onstrated that NH OH oxidation is coupled stoichiometrically NO2 after 20 h of incubation. Data are means of triplicates, with variation 2 less than 10%. The experiment was repeated at least three times and pro- (1:1) to reduction of O2 (see above), suggesting electron transfer duced similar results. Error bars represent SEM. from NH2OH to O2. Given the limited N. maritimus biomass

2of6 | www.pnas.org/cgi/doi/10.1073/pnas.1214272110 Vajrala et al. Downloaded by guest on September 25, 2021 15 available, direct measurement of electron transport or proton oxidation of then significantly N-enriched NH2OH pool con- 15 − translocation in response to NH2OH was impossible. Thus, to tinued, leading to further accumulation of NO2 at a lower rate confirm the metabolic coupling of NH2OH oxidation, we mea- (Fig. 4 A and B). Due to rapid decomposition of NH2OH in SCM − sured ATP levels and NO2 production rates in the presence of media, we could not establish the mass and isotope balance + − 15 different substrates. Cellular ATP levels were measured follow- among NH4 ,NH2OH, and NO2 in our N experiments. + ing addition of SCM media containing either no substrate, NH4 , Moreover, presence of high concentrations of NH2OH was toxic N. maritimus + or NH2OH to membrane-immobilized N. maritimus cells. Cells to and slowed down the rates of NH4 oxidation − incubated in the presence of NH2OH showed statistically sig- and NO2 production. 15 nificantly higher levels of ATP compared with cells incubated Because the determination of δ N signatures of NH2OH and − without substrate (P value 0.006). Furthermore, cells incubated NO2 entailed a chemical conversion step affording the more with NH2OH showed statistically significantly higher ATP levels stable N2O for subsequent stable isotope analysis, a series of + 15 compared with cells incubated with NH4 (P value 0.007) (Fig. controls were conducted to exclude other sources of N accu- − + + 3A) although the NO2 produced in presence of either NH4 or mulation in the N2O pool. Neither NH4 itself, nor N2H4 were NH OH is almost similar (Fig. 3B). Addition of either carbonyl found to yield N2O by the chemical conversion method used 2 + cyanide m-chlorophenyl hydrazone (CCCP), a well-characterized here, thus, ruling out a possible interference by NH4 itself or uncoupler of ATP synthesis, or use of heat-killed cells, prevented a role of N2H4 in the N. maritimus metabolism. We further N. maritimus NH3-dependent or NH2OH-dependent increase in ATP levels in eliminated the possibility of generation of N2Oin all treatments (Fig. 3A). Furthermore, CCCP addition diminished cell suspensions from chemical decomposition of unstable N − − NO production from NH , but not NO production from in- species, such as nitric oxide (NO), nitroxyl (HNO), or nitrosyl 2 3 2 + cubation with NH2OH (Fig. 3B). These results suggest that the (NO ), or their rapid reaction with added NH2OH. Although respiratory reduction of O2 associated with NH2OH oxidation is such chemical reactions cannot be completely excluded, their coupledtoATPproductioninN. maritimus, as has been reported significance during our experiments must be negligible because − in N. europaea (24). Inhibition of NO2 production from NH3 by N2O concentrations in the control vials did not increase over CCCP further suggests that analogous to the AOB, the oxidation time, indicating no significant new production of N2O through of NH3 in N. maritimus by AMO also depends on input of elec- either of these reactions. Thus, although NH2OH does not ac- trons from electron carriers (7). Consistent with this dual sink of cumulate at measurable concentrations during normal growth of

N. maritimus SCIENCES electrons in the NH3 oxidation metabolism, the cellular level of in culture, production and consumption of NH2OH + ENVIRONMENTAL ATP is lower during incubation with NH4 compared with incu- during oxidation of NH3 remains the only plausible explanation bations with NH2OH only (Fig. 3A). for our observed results, indicating that NH2OH is an intermediate in the NH3 oxidation pathway of N. maritimus. Generation of NH2OH from NH3. We used stable isotope tracer experiments to further establish that NH2OH is an intermediate in Discussion the NH3 oxidation pathway of N. maritimus. Membrane-immo- The lack of an identifiable gene encoding HAO in the genome of bilized cells were incubated with natural abundance NH4Cl and N. maritimus (8) necessitates either a novel enzyme for the oxi- μ 15 NH2OH (200 M each). The N isotopic composition of NH2OH dation of NH2OH or, alternatively, an initial product of NH3 − and accumulated NO2 were measured over time by isotope ratio oxidation other than NH2OH (8). We earlier speculated that the mass spectrometry (IR-MS) following chemical conversion of immediate oxidation product of the archaeal variant of AMO − NH2OH or NO2 to nitrous oxide (N2O) by iron acetate (FeAc) or might be a reactive N species, such as nitroxyl, instead of NH2OH sodium azide, respectively (25). NH2OH-derived N2O and N2O (8). However, the data presented in this study now clearly im- present in the control vials before chemical conversion of NH2OH plicate NH2OH as the immediate product of AMO, with prop- to N2O were quantified by gas chromatography (GC). During 60 erties similar to those previously observed in N. europaea (20). − − min of incubation, NO2 concentrations steadily increased and NO2 accumulation and O2 uptake during NH2OH oxidation by NH2OH concentrations decreased (Fig. 4 A and B). After addition N. maritimus were stoichiometrically coupled (Fig. 2) and closely 15 15 of 20 μM NH4Cl, the relative abundance of N increased over paralleled NH2OH disappearance and therefore fulfilled the − time in the NH2OH and NO2 pools (Fig. 4 A and B). After 120 min, expected 1:1 stoichiometry of NH2OH and O2 consumption. NH3 oxidizing activity was blocked by addition of C2H2. Accu- Metabolism of NH2OH was not inhibited by C2H2 or ATU, two 15 mulation of N stopped completely in the NH2OH pool within well-characterized inhibitors of the bacterial AMO. Both archaeal 20 min of C2H2 addition, indicating complete inhibition of me- and bacterial NH3 oxidation are inhibited by C2H2, a mechanism- 15 tabolism of highly N-labeled NH3. In contrast, uptake and based inactivator of NH3 oxidation in the AOB (Fig. S1A) (21, 26).

− Fig. 3. ATP formation by N. maritimus.(A) ATP and (B)NO2 concentration in N. maritimus cells incubated in presence of either no substrate (white bars), 200 + − μMNH4 (gray bars), or 200 μMNH2OH (black bars). Concentration of ATP and NO2 in heat-killed cells and cells treated with 100 μM CCCP for 20 min is also − − shown. The NO2 trace amounts detected in the controls are likely due to leaching of residual traces of NO2 in the nylon ﬁlters used to harvest N. maritimus cells. Error bars represent the SEM.

Vajrala et al. PNAS Early Edition | 3of6 Downloaded by guest on September 25, 2021 15 Fig. 4. Kinetics of N. maritimus NH2OH consumption and accumulation of N in a stable isotope tracer experiment. N. maritimus cells were incubated in the 15 presence of 200 μMNH4Cl and 200 μMNH2OH. After 60 min (arrow) 20 μM NH4Cl (99.5% purity) was added to the cell suspension. After 120 min NH3- 15 oxidation activity was blocked by addition of C2H2 (0.1%). (A) Consumption of NH2OH (dashed lines, filled circles) and accumulation of NinNH2OH pool − 15 − (solid lines, filled diamonds). (B) Accumulation of NO2 (dashed lines, filled circles), NinNO2 pools (solid lines, filled diamonds).

As observed in AOB, C2H2 did not inactivate NH2OH oxidation in it neither increased nor decreased significantly over time. Albeit, we N. maritimus (Fig. S1A). Previously, C2H2 was shown to be a sui- cannot state its isotopic composition with confidence, the measured cidal substrate for AMO in N. europaea and that protein synthesis δ15N of unconverted and converted samples were in a similar range was essential for recovery from C2H2 inactivation (19, 26). N. throughout the experiments (−28.8 to + 66.3, and −32.7 to + 140.0 15 maritimus cells inactivated by C2H2 also require a recovery period, for NH2OH-derived and direct N2O, respectively). Thus, Nac- 15 not showing any respiratory capacity until 60–90 min following cumulating in the N2O pool contributed a very minor fraction of N + NH4 addition. In addition, the oxidation of NH2OH by N. mar- accumulating in the NH2OH pool. Previous studies demonstrate itimus remained virtually unaffected by concentrations of ATU, an that N2O production during NH3 oxidation in enrichment culture inhibitor of the bacterial AMO, that completely inhibits AOB and from oceanic water samples, enrichments, and N. maritimus only − AOA NH3 oxidation at the concentrations used in our experi- contribute up to two parts per thousand of NO2 produced (29). A ments (Fig. S1B) (20). Thus, although the enzymes and metabolic similar rate of N2O production by N. maritimus cells during our pathway of AOA have not yet been characterized (8–12), we have experiments would yield an estimated 70 nM N2O over the course now shown that N. maritimus is capable of oxidizing NH2OH in of our entire experiments. Because the N2Oconcentrationsinthe presence of C2H2 similar to N. europaea. Thus, we can also discard unconverted samples were below detectable levels, we conclude the possibility that C2H2 interferes with components other than the that any direct N2O produced in our experiments would not affect AMO protein in the AOA. the NH2OH isotopic analysis. − Having demonstrated that NH2OH can be oxidized to NO2 The isotopic signature of N2O produced by AOA strongly by N. maritimus, we then considered whether this oxidation indicated that it is a product of NH3 oxidation and not from − provided a benefit to the cells. This is a necessary consequence of reduction of NO2 (28). In addition, C2H2 (∼10%) is known to NH2OH serving as an intermediate in NH3 oxidation, because all inhibit N2O reductase but not nitrite reductase (NirK) or nitrous electrons for respiration are derived from the oxidation of oxide reductase (NOR) activities of nitrifier denitrification path- 15 NH2OH. Indeed, a benefit was observed by marked ATP pro- ways in AOB (30, 31). In our N tracer experiments, the decline 15 duction in response to NH2OH addition relative to resting cells of NH2OH-derived N2O accumulation after addition of 0.1% 15 without substrate (Fig. 3A). Similar, but lower, ATP production C2H2, indicated that the N2O is derived from oxidation of 15 15 − coupled to oxidation of NH2OH was measured in N. europaea NH3 and not from reduction of NO2 (Fig. 4A). We also note (24, 27). Although we have no explanation for this differential the formal possibility that exogenously added NH2OH could be response, N. maritimus may maintain a slightly higher energy converted chemically or biologically forming short-lived nitrogen • • charge than that characterized for AOB. radicals, such as nitroxyl (HNO ), nitrogen monoxide (NO ), or − N. + Direct evidence that the oxidation of NH3 to NO2 by nitrosyl (NO ), all of which under O2 limitation could undergo maritimus proceeds via NH2OH came from stable isotope tracer further reactions yielding N2O to varying degrees (32). N2O 15 experiments in which NH2OH production was detected while generation can also result from biological or chemical conversion N. maritimus 15 A − was oxidizing NH3 (Fig. 4 ). One possible caveat of NO2 and NH2OH into N2O (33). However, a role of such 15 to this interpretation is that accumulation of NH2OH was mea- cross-reactions involving NH2OH itself as well as nitrogen-free 15 sured by IR-MS after its conversion to N2O. Recent reports radicals can also be ruled out because addition of C2H2 stops 15 demonstrated that archaeal NH3 oxidation produces N2O (28) further accumulation of N2O in total N2O pool. Together, albeit no biochemical pathway for its production in AOA has these results clearly demonstrate that NH2OH is a product of fi N. maritimus been identi ed (29). Production of N2O either by NH3 oxidation in N. maritimus. itself or chemical reactions affording N2O in our growth media, We are thus left with a unique enzyme system for the oxidation e.g., through decomposition of NH2OH, could potentially yield of NH2OH in the AOA. All data now implicate a variety of N2O that would interfere with NH2OH isotopic analysis. We unusual proteins, rather than the c-type cytochrome system of attempted to quantify N2O concentrations and isotopic compo- NH2OH oxidation and respiration in the AOB, comprising the sition in parallel in iron acetate-converted and -unconverted sam- remaining oxidative and energy-harvesting steps in the AOA. ples under identical conditions. However, N2O concentrations in Apart from providing additional support for the role of a unique the unconverted samples were too low and outside the calibration biochemistry in global nitrification, continued characterization of range of our IR-MS setup, even after increasing the injection vol- this pathway may point to environmental factors, such as specific ume 10-fold. We estimate the highest N2O concentration in any trace metal availability, limiting or promoting nitrification in unconverted sample was 6–10 nM (median 7.75, SD 1.6, n = 9) and different marine and terrestrial provinces.

4of6 | www.pnas.org/cgi/doi/10.1073/pnas.1214272110 Vajrala et al. Downloaded by guest on September 25, 2021 Methods ATP Measurements. For ATP measurements, the N. maritimus cultures were grown to midlog phase and filtered using the manifold setup. The filters Culture Conditions and Cell Harvesting. All physiological experiments were + carried out with N. maritimus strain SCM1 in Hepes-buffered synthetic carrying the N. maritimus cells were immersed in NH4 -free SCM in a serum vial capped with a gray-butyl rubber stopper. The bottles were purged with crenarchaeota medium (SCM) prepared as described (16). C2H2 was pro- argon gas and cells were allowed to incubate for 2 h to deplete their en- duced from barium carbide (BaC2) as described by Hyman and Arp (34). dogenous ATP levels, allowing detection of newly synthesized ATP upon Other reagents, such as NH4Cl, NH2OH, and ATU were research-grade addition of the substrates. After 2 h of anaerobic treatment, identical filter products and were obtained from Sigma. N. maritimus cells were grown to − pieces were cut and distributed into wells of a white 96-well assay plates midlog phase (∼550–650 μMNO accumulated) and were concentrated by 2 (White w/lid, tissue culture-treated; BD Biosciences). The plate wells con- filtration onto nylon membrane filters (Whatman; 47 mm diameter, 0.2 μm tained O -saturated SCM with (i) no substrate, (ii) 200 μMNH+,or(iii) 200 pore size; 7402-004) using a vacuum manifold system at 600 mm Hg pres- 2 4 μMNHOH. ATP accumulation was measured over a time course using sure. The N. maritimus cells harvested in this fashion actively metabolized 2 a commercial kit based on the luciferase enzyme activity (BactTiter Glo; NH for at least 8 h, enabling short-term assays to be conducted. Multiple 3 Promega). To detect ATP generation, 100 μL of the reagent was added per liters of N. maritimus cultures could be filtered simultaneously and the nylon well, incubated 5 min, and the bioluminescence was measured with a mul- membranes with concentrated cells served as solid support for the experi- tifunction plate reader (Infinite M200; Tecan) with 1-s integration and 10-ms mental incubations. The N. maritimus cells on the nylon membranes were + + settle time. Statistical differences in ATP values between NH OH and NH or incubated in SCM with either NH or NH OH added as a substrate. 2 4 4 2 no substrate were evaluated using Student’s t test. An appropriate ATP standard curve was used to estimate the concentration of ATP in the sam- NH3- and NH2OH-Dependent O2 Uptake Measurements. Rates of NH3 and − ples. Samples for NO2 determination were collected before addition of ATP NH OH-dependent O uptake of the N. maritimus cells adsorbed onto the 2 2 reagent and analyzed as described above. nylon membrane filters were measured with a Clark-type O2 electrode (Yellow Springs Instrument) mounted in an all-glass, water-jacketed reaction Detection of NH OH Formation in N. maritimus Cells. Natural abundance NH Cl vessel (18 mL volume) and operated at 30 °C. All measurements were carried 2 4 + + and NH OH (200 μM each) were added to SCM containing N. maritimus cells out in NH -free SCM media with added NH or NH OH (200 μM each). 2 4 4 2 adsorbed onto the nylon membrane filters. The bottles were incubated in

− the dark at 30 °C. Samples were taken at 20-min intervals for determination Determination of NH2OH, NO2 , N2O and Protein Concentration. NH2OH in SCM − − of NO2 ,N2O concentrations, and isotopic compositions of NH2OH and NO2 . was oxidized to nitrous oxide (N2O) by the addition of Fe (III) as described 15 15 A total of 20 μM NH4Cl (Cambridge Isotopes; 99 atom % N) was added at (25) and with a minor modification. In our study, the FeAc reagent consisted − 60-min time point and 0.1% C2H2 was added at 120-min time point. NO2 of 8% (wt/vol) colloidal iron oxide and 6% (vol/vol) glacial acetic acid. This concentration was determined by Griess reagent as described above. modification adjusted the pH of the sample below 1.4 and assured complete Total NH2OH was determined by GC-thermal conductivity detector (TCD) SCIENCES conversion of NH2OH to N2O. The FeAc reagent (1 mL) was injected into 20 after conversion to N O as described above. Isotopic composition of N Owas fl 2 2 ENVIRONMENTAL mL amber glass serum bottles sealed with Te on caps containing 2 mL of determined using a continuous flow ThermoFinnigan DeltaPlus stable iso- sample and the bottles were mixed thoroughly. After 1 h incubation at 30 ° tope ratio mass spectrometer (Thermo) coupled to a Precon (Thermo) and C, saturated NaCl solution (1 mL) was injected into the sealed bottles to − Gasbench II (Thermo) relative to N2O and NO3 standards as described pre- displace the N2O in the liquid phase into the headspace. The bottles were 15 14 15 14 viously (36, 37). Even though the δ N/ NinN2O pools and the AT% N/ N then flash frozen on dry ice and stored upside down until further analysis. increased during incubations, the rapid disappearance of NH2OH caused the The N2O gas in the headspace was measured by gas chromatography using 15 15 μM N-NH2OH concentrations to decrease over time. Hence μM N-NH2OH a thermal conductivity detector. Samples were analyzed before and after concentrations were calculated from AT % excess 15N/14N values. The AT % incubation with FeAc reagent to determine if any N2O accumulated during 15 15 excess N and μM concentrations of NH2OH are calculated as defined below: NH2OH oxidation before incubation with FeAc reagent. The conversion of 15 = 15 NH2OH to N2O proceeded in a 2:1 stoichiometry and was above 95% in i)AT% N excess in a sample AT % N in labeled NH2OH pool of sample + − – 15 a concentration range of 1 μMto1mMNH2OH. Neither NH4 nor NO2 AT % natural abundance N in unlabeled NH2OH pool at the beginning interfered with the conversion and quantification of NH2OH. of the experiment; and − 15 15 ii) μM N-NH2OH = (μM total NH2OH)/(AT % N excess in a sample × 100). NO2 concentration was determined with the Griess reagent (sulfanilamide − and N-naphthylethylenediamine) as described (35). The rate of NO2 accumulation has been shown previously to be proportional to the density of ACKNOWLEDGMENTS. We thank David Myrold for valuable discussion on N. maritimus cell culture (16) and was used as an indicator of growth stages. the stable isotope tracer experiment results. This work was funded by the US fi Protein concentrations of N. maritimus cells adsorbed onto the lter were Department of Agriculture (Grant 2005-35319), by the Oregon Agricultural determined with the micro BCA protein assay kit from Pierce after protein Experiment Station, and by the National Science Foundation (MCB-0604448 solubilization at 60 °C for 60 min. and MCB-0920741).

1. de la Torre JR, Walker CB, Ingalls AE, Könneke M, Stahl DA (2008) Cultivation of 11. Nunoura T, et al. (2011) Insights into the evolution of Archaea and eukaryotic protein a thermophilic ammonia oxidizing archaeon synthesizing crenarchaeol. Environ Mi- modifier systems revealed by the genome of a novel archaeal group. Nucleic Acids Res crobiol 10(3):810–818. 39(8):3204–3223. 2. Hatzenpichler R, et al. (2008) A moderately thermophilic ammonia-oxidizing cren- 12. Spang A, et al. (2010) Distinct gene set in two different lineages of ammonia-oxidizing archaeote from a hot spring. Proc Natl Acad Sci USA 105(6):2134–2139. archaea supports the phylum Thaumarchaeota. Trends Microbiol 18(8):331–340. 3. Könneke M, et al. (2005) Isolation of an autotrophic ammonia-oxidizing marine ar- 13. Kim BK, et al. (2011) Genome sequence of an ammonia-oxidizing soil archaeon, “ ” – chaeon. Nature 437(7058):543–546. Candidatus Nitrosoarchaeum koreensis MY1. J Bacteriol 193(19):5539 5540. 4. Lehtovirta-Morley LE, Stoecker K, Vilcinskas A, Prosser JI, Nicol GW (2011) Cultivation 14. Mosier AC, Allen EE, Kim M, Ferriera S, Francis CA (2012) Genome sequence of “ ” of an obligate acidophilic ammonia oxidizer from a nitrifying acid soil. Proc Natl Acad Candidatus Nitrosopumilus salaria BD31, an ammonia-oxidizing archaeon from the San Francisco Bay estuary. J Bacteriol 194(8):2121–2122. Sci USA 108(38):15892–15897. 15. Mosier AC, Allen EE, Kim M, Ferriera S, Francis CA (2012) Genome sequence of 5. Tourna M, et al. (2011) Nitrososphaera viennensis, an ammonia oxidizing archaeon “Candidatus Nitrosoarchaeum limnia” BG20, a low-salinity ammonia-oxidizing ar- from soil. Proc Natl Acad Sci USA 108(20):8420–8425. chaeon from the San Francisco Bay estuary. J Bacteriol 194(8):2119–2120. 6. Arp DJCPS, Chain PS, Klotz MG (2007) The impact of genome analyses on our un- 16. Martens-Habbena W, Berube PM, Urakawa H, de la Torre JR, Stahl DA (2009) Am- derstanding of ammonia-oxidizing bacteria. Annu Rev Microbiol 61:503–528. monia oxidation kinetics determine niche separation of nitrifying Archaea and Bac- 7. Arp DJ, Sayavedra-Soto LA, Hommes NG (2002) Molecular biology and biochemistry teria. Nature 461(7266):976–979. of ammonia oxidation by Nitrosomonas europaea. Arch Microbiol 178(4):250– 17. Lees H (1952) Hydroxylamine as in intermediate in nitrification. Nature 169:156–157. 255. 18. de Bruijn P, Van de Graaf AA, Jetten MSM, Robertson LA, Kuenen JG (1995) 8. Walker CB, et al. (2010) Nitrosopumilus maritimus genome reveals unique mecha- Growth of Nitrosomonas europaea on hydroxylamine. FEMS Microbiol Lett 125(2–3): fi nisms for nitri cation and autotrophy in globally distributed marine crenarchaea. 179–184. Proc Natl Acad Sci USA 107(19):8818–8823. 19. Hyman MR, Wood PM (1985) Suicidal inactivation and labelling of ammonia mono- 9. Hallam SJ, et al. (2006) Genomic analysis of the uncultivated marine crenarchaeote oxygenase by acetylene. Biochem J 227(3):719–725. Cenarchaeum symbiosum. Proc Natl Acad Sci USA 103(48):18296–18301. 20. Hofman T, Lees H (1953) The biochemistry of the nitrifying organisms. IV. The res- 10. Blainey PC, Mosier AC, Potanina A, Francis CA, Quake SR (2011) Genome of a low- piration and intermediary metabolism of Nitrosomonas. Biochem J 54(4):579–583. salinity ammonia-oxidizing archaeon determined by single-cell and metagenomic 21. Offre P, Prosser JI, Nicol GW (2009) Growth of ammonia-oxidizing archaea in soil analysis. PLoS ONE 6(2):e16626. microcosms is inhibited by acetylene. FEMS Microbiol Ecol 70(1):99–108.

Vajrala et al. PNAS Early Edition | 5of6 Downloaded by guest on September 25, 2021 22. Böttcher S, Koops H-P (1994) Growth of lithotrophic ammonia-oxidizing bacteria on 30. Berg P, L Klemedtsson, Rosswall T (1982) Inhibitory effect of low partial pressures of hydroxylamine. FEMS Microbiol Lett 122:263–266. acetylene on nitrification. Soil Biol Biochem 14(3):301–303. 23. Basant Bhandari DJDN (1979) Ammonia and O2 uptake in relation to proton trans- 31. Bateman EJ, Baggs EM (2005) Contributions of nitrification and denitrification to location in cells of Nitrosomonas europaea. Arch Microbiol 122(3):249–255. N2O emissions from soils at different water-filled pore space. Biol Fertil Soils 41(6): 24. Ramaiah A, Nicholas DJ (1964) The synthesis of Atp and the incorporation of 32p by cell- 379–388. free preparations from Nitrosomonas europaea. Biochim Biophys Acta 86:459–465. 32. Moir JW, Crossman LC, Spiro S, Richardson DJ (1996) The purification of ammonia 25. Butler JH, Gordon LI (1986) Rates of nitrous oxide production in the oxidation of monooxygenase from Paracoccus denitrificans. FEBS Lett 387(1):71–74. hydroxylamine by iron(II1). Inorg Chem 25:4573–4577. 33. Hooper AB, Terry KR (1979) Hydroxylamine oxidoreductase of Nitrosomonas. Pro- 26. Hyman MR, Arp DJ (1992) 14C2H2- and 14CO2-labeling studies of the de novo syn- duction of nitric oxide from hydroxylamine. Biochim Biophys Acta 571(1):12–20. thesis of polypeptides by Nitrosomonas europaea during recovery from acetylene and 34. Hyman MR, Arp DJ (1990) The small-scale production of [U-14C]acetylene from light inactivation of ammonia monooxygenase. J Biol Chem 267(3):1534–1545. Ba14CO3: Application to labeling of ammonia monooxygenase in autotrophic nitri- 27. Frijlink MJ, Abee T, Laanbroek HJ, de Boer W, Konings WN (1992) The bioenergetics fying bacteria. Anal Biochem 190(2):348–353. of ammonia and hydroxylamine oxidation in Nitrosomonas europaea at acid and 35. Hageman RH, Hucklesby DP (1971) Nitrate reductase in higher plants. Methods En- alkaline pH. Arch Microbiol 157(2):194–199. zymol 23:491–503. 28. Santoro AE, Buchwald C, McIlvin MR, Casciotti KL (2011) Isotopic signature of N(2)O 36. Sansone FJ, Popp BN, Rust TM (1997) Stable carbon isotopic analysis of low-level produced by marine ammonia-oxidizing archaea. Science 333(6047):1282–1285. methane in water and gas. Anal Chem (69):40–44. 29. Löscher CRKA, Könneke M, LaRoche J, Bange HW, Schmitz HW (2012) Production of 37. Sigman DM, et al. (2001) A bacterial method for the nitrogen isotopic analysis of oceanic nitrous oxide by ammonia-oxidizing archaea. Biogeosciences 9:2419–2429. nitrate in seawater and freshwater. Anal Chem 73(17):4145–4153.

6of6 | www.pnas.org/cgi/doi/10.1073/pnas.1214272110 Vajrala et al. Downloaded by guest on September 25, 2021