Although different in their basic modes of metabolism, methane-oxidizing (methanotrophic) bacteria and obligate lithotrophic ammonia-oxidizing bacteria (AOB) share physiological characteristics that arise mainly from their ability to initiate the oxidation of both methane (CH4) and ammonia (NH3) with homologous enzymes, particulate methane monooxygenase (pMMO) and ammonia monooxygenase (AMO), respectively (Holmes et al., 1995; Klotz and Norton, 1998; Norton et al., 2002). On the basis of physiological studies (Jones and Morita, 1983; Ward, 1987, 1990) and genomic inventories (Ward et al., 2004; Arp et al., 2007), methanotrophs and AOB use distinct pathways for the assimilation of carbon and production of CO2 from CH4. On the other hand, their pathways for the oxidation of NH3 to nitrite (NO2) and possibly the production of N-oxides directly or as by-products of NH3 catabolism appear related in terms of mechanism and genomic inventory (Zahn et al., 1994; Arp and Stein, 2003; Hooper et al., 2005; Klotz and Stein, 2008). An understanding of the latter process is particularly important for the understanding of the global nitrogen cycle, as recent studies indicate that both groups of bacteria may release considerable amounts of N2O (Klotz and Stein, 2008; and references therein).

Pioneering work by Howard Dalton and associates showed that Methylococcus capsulatus Bath, like AOB, produces hydroxylamine (NH2OH) as a product of NH3 oxidation, and that this conversion is facilitated by pMMO or the soluble MMO (Dalton, 1977). More than a decade later, a pathway for the oxidation of NH3 to NO2 for M. capsulatus Bath was finally described (Zahn et al., 1994; Bergmann et al., 1998). However, AOB and M. capsulatus Bath were thought to differ in the enzymes used for the intermediate conversion of NH2OH to NO2 (DiSpirito et al., 2005; Hooper et al., 2005; and references therein). Known AOB genomes encode two enzymes that oxidize NH2OH, hydroxylamine oxidoreductase (HAO) and cytochrome P460 (Arp and Stein, 2003; Hooper et al., 2005; Arp et al., 2007). HAO oxidation of NH2OH provides electrons for both AMO activity and the generation of proton motive force; HAO exhibits much higher NH2OH-oxidizing activity than cytochrome P460 (Hooper et al., 2005; and references therein). Because AOB cannot grow without functional HAO (Hommes et al., 2002), the lower activity of cytochrome P460 is consistent with a physiological role as an ancillary N-oxide detoxification enzyme. However, the role of cytochrome P460 might differ in M. capsulatus Bath, as it is not an obligate lithotrophic ammonia oxidizer (Klotz and Stein, 2008).

Before its genome sequence was known, M. capsulatus Bath was assumed to lack HAO and to use a cytochrome P460 in its nitrification pathway (Zahn et al., 1994). The hao gene is transcribed in response to NH3 in the AOB Nitrosomonas europaea (Sayavedra-Soto et al., 1996), whereas in M. capsulatus Bath, cytochrome P460 expression is low in either the presence or the absence of NH3 (Bergmann et al., 1998, 2000). Determination of the genome sequence of M. capsulatus Bath revealed a putative homolog of the HAO protein (Ward et al., 2004; Bergmann et al., 2005). This protein not only shares more than 65% sequence similarity with Hao from AOB, but also contains the heme and ligand-coordinating residues (Tyr467, Aps267 and His268) that are critical to the function of the trimeric HAO complex (Bergmann et al., 2005). Moreover, a conserved open reading frame (orf2) of unknown function is located 3′ of the hao gene in M. capsulatus Bath and all known AOB (Bergmann et al., 2005; Arp et al., 2007). This conservation of primary and secondary Hao structure as well as observed synteny suggest that M. capsulatus Bath could produce an active HAO trimer. If M. capsulatus Bath transcribes hao and uses HAO for NH2OH oxidation, then the nitrification pathways of methanotrophs and AOB could be more similar than has been appreciated previously.

The objectives of this study then were (1) to determine whether M. capsulatus Bath transcribes hao and orf2, (2) to assess the production of these gene transcripts in response to NH3 and (3) to quantify transcription of mRNAs for other genes encoding products (cytochrome P460, cytochrome c′-β (CytS) and nitric oxide (NO) reductase cNOR) that may function in the detoxification of N-oxides produced during NH3 catabolism.

Results and discussion

Use of quantitative PCR (qPCR) for the analysis of M. capsulatus Bath gene expression

As we did not find any prior descriptions of the use of qPCR to study M. capsulatus Bath gene expression, we first determined the expression of transcript levels for pmoB (encoding the catalytic subunit of pMMO (Prior and Dalton, 1985; Martinho et al., 2007)), cytL (cytochrome P460) and cytS for comparison with previous determinations by northern blots (Nielsen et al., 1997; Bergmann et al., 2000). We used the Relative Expression Software Tool (REST v.2005) (Pfaffl et al., 2002) to calculate expression ratios, which represent the n-fold differences in the levels of a target sequence in the CH4 plus 5 mM (NH4)2SO4 (CH4+NH3) treatment compared to the CH4-only treatment for all genes tested. As expected, pmoB was expressed at constant basal levels in both the CH4-only and CH4+NH3 treatments when expression ratios were normalized to 16S rRNA (rrsA) in independent experiments. Because pmoB transcript levels did not vary between treatments, we used this mRNA along with rrsA as an internal standard in comparing transcript levels for other genes. Also as expected, we found that cytL transcription did not respond to NH3, but observed a strong increase (28.5- and 40-fold) in cytS transcript levels in the presence of NH3 (Figure 1b). Both results are consistent with previous reports (Bergmann et al., 2000). Because qPCR yielded expected results for previously studied genes with different predicted expression patterns, we deemed it suitable as a method to study the transcript levels for genes whose expression had not been described previously (hao, orf2, norC and norB).

Figure 1
figure 1

Map of the haoAB (formerly hao–orf2, Bergmann et al., 2005) gene cluster from Methylococcus capsulatus Bath (a) and expression ratios calculated from Ct values of products amplified with primers indicated in the figure and listed in Table 1 (b). (a) Map of the haoAB gene cluster with locations of the putative promoter (flag) for haoA, stem-loop structures and amplicon positions. The translational start codon (TTG) of haoA (locus tag: MCA0956) beginning at position 1000257 (complement strand) in the NCBI annotation (AE017282) has been relocated to position 1000182 based on the presence of σ70 promoter consensus sequences and a ribosomal binding site (RBS). The first and second stem-loop structures have ΔG-values of −25.3 and −16.5 kcal mol−1, respectively, but only the second terminator is succeeded by a stretch of thymidines. (b) Expression ratios were calculated with the REST program using Ct values of products amplified with primers listed in Table 1. Black and gray bars represent independent trials 1 and 2, respectively, for each primer set. Error bars represent standard errors calculated for triplicate qPCR reactions. The horizontal line demarcates an expression ratio of 2. Letters designate significance (a=P<0.05; b=nonsignificant) of the increased expression ratio as tested through the randomization of the data set. Ct, cycle threshold; REST, relative expression software tool; qPCR, quantitative PCR.

Determination of hao and orf2 transcript levels

To determine whether M. capsulatus Bath expresses hao in response to NH3 and cotranscribes hao and the adjacent downstream orf2, we performed qPCR on cDNA synthesized from total RNA of duplicate cultures exposed either to CH4 or to CH4 plus 5 mM (NH4)2SO4. Primer sets (Table 1; Figure 1a) were designed to amplify the 5′ and 3′ ends of hao (q1hao and q2hao, respectively), as well as to target the overlap between these genes (qhaoorf2) and two regions within orf2 (q1orf2 and q2orf2). The fluorescence versus cycle number curves and electrophoretic verification of product size (data not shown) indicated that all primer sets amplified the expected target sequences from cDNA. No products were obtained from the corresponding nonreverse-transcribed RNA samples (negative control), indicating that M. capsulatus Bath transcribed hao and orf2 in the presence of either CH4 or CH4+NH3.

Table 1 Primer sets used in qPCR of cDNA from M. capsulatus Bath

Transcription of hao by M. capsulatus Bath exposed to CH4+NH3 was 14.5- and 31-fold (P<0.05) higher than under CH4-only conditions based on the relative expression ratios obtained for the two independent trials, when calculated with the cycle threshold (Ct) values of the q1hao amplicon located at the 5′ end of the transcript (Figure 1a). The q2hao primer set targeting the 3′ end of hao showed that transcription increased only sixfold (P<0.05) in response to NH3, which reflects an apparent decrease in steady-state levels of hao and orf2 mRNA around a putative stem-loop structure (Figure 1a). The qhaoorf2 primer set targeting a region that spans the 3′ end of hao and 5′ end of orf2 (Figure 1a) yielded expression ratios consistent with those obtained from qPCR with the q2hao primer set (Figure 1b). On the basis of both primer pairs targeting orf2 (q1orf2, q2orf2), cDNA from one trial showed that M. capsulatus Bath exposed to CH4+NH3 increased the expression of orf2 by 9-fold and 12-fold (P<0.05), whereas the other cDNA set yielded expression ratios of 3.2 for both amplified regions of orf2, but with marginally nonsignificant P-values of 0.097 (q1orf2) and 0.083 (q2orf2) for the randomization test performed by the REST program (Figure 1b). The Ct values of hao and orf2 targets amplified from cDNA were higher in the CH4-only cultures than in CH4+NH3 cultures (Supplementary Table S1; note that Ct values are inversely proportional to the expression levels).

A comparison of Ct values obtained with each primer set on cDNA derived from both CH4-only and CH4+NH3 cultures as described above showed variation in the steady-state mRNA levels in different regions along the hao and orf2 transcript, with primers designed to detect the upstream hao yielding smaller Ct values (higher steady-state levels) than those targeting orf2 and the overlapping region between the two ORFs. There are several possible explanations for differences seen when targeting different regions with the same transcript. Given the higher G+C content of orf2 (70%) compared with hao (62%), the results could be a methodological artifact due to different efficiencies of in vitro reverse transcription of RNA. On the other hand, there is a putative stem-loop structure (ΔG=−25.3 kcal mol−1) centered approximately 29 base pairs downstream from the putative translational start of orf2 (Figure 1a) possibly capable of forming a Rho-independent terminator, and resulting in reduced orf2 transcript levels relative to hao. Although other factors (that is, alternate secondary structures arising during elongation or degradation, the presence of ribosomal complexes, proteins and other RNA species that interact with nascent RNA chains and so on) can also influence the steady-state levels of mRNA in vivo (Carrier and Keasling, 1997), our data are consistent with a post-transcriptional regulatory mechanism in synthesis of the hao–orf2 transcript. A similar situation was found for transcription of the gene cluster encoding AMO in the AOB Nitrosococcus oceani. This transcriptional unit of 5 genes (amoRCABD) was reported to contain leaky terminators (between amoC and amoA as well as amoB and amoD) and premature termination at these leaky terminators accounted for the difference in transcript concentrations as measured by northern analysis (El Sheikh and Klotz, 2008; El Sheikh et al., 2008).

M. capsulatus Bath cotranscribes hao and orf2 in response to ammonia and no sequences similar to known promoter elements upstream of orf2 were identified. As there is conservation of haoorf2 synteny in all known AOB and M. capsulatus Bath, we propose that these genes be named haoA and haoB, respectively, and we refer to them as such in the remainder of this paper.

Transcription of norC and norB and implications for NO reduction

M. capsulatus Bath produces N2O when supplied with either NH3 or NH2OH (Sutka et al., 2003, 2006), although the pathway is not well characterized. While NO production has not been detected in M. capsulatus Bath, presumably it is a precursor of N2O production, and both NO and N2O are presumed to be generated as by-products of NH2OH oxidation by cytochrome P460 (Sutka et al., 2006). The presence of putative NO reductase genes (norC and norB) in the genome of M. capsulatus Bath raises the possibility that N2O production is catalyzed by a dedicated enzyme. In many bacteria, genes for NO reductase are expressed in response to NO and nitrosating agents (sodium nitroprusside and S-nitrosothiols and so on) (Poole, 2005). Hence, we designed primer pairs (Table 1, qnorC, qnorCB and qnorB) for the contiguous genes encoding NO reductase subunits C and B (norC and norB) of M. capsulatus Bath to test whether these genes are expressed in the presence of ammonia. All primer sets amplified products from CH4-only and CH4+NH3 cDNA (data not shown). These genes were cotranscribed and produced similar steady-state levels of mRNA in the CH4-only and CH4+NH3 treatments (Figure 1b). Hence, norCB transcription did not respond to the presence of ammonia.

Cytochrome c NO reductase (cNOR) and CytS (Elmore et al., 2007) are candidate enzymes for mediating NO reduction to N2O in M. capsulatus Bath; however, the genomic inventory involved in NO production remains elusive. The typical mechanism for NO reduction involves nitrite reductases (Ferguson and Richardson, 2005), yet there are neither known NO-producing (NirK and NirS) nor ammonium-producing (NrfA and NirB) nitrite reductases in the genome of M. capsulatus Bath (Ward et al., 2004). In addition to copper-dependent nitrite reductase (NirK), HAO has also been implicated in NO production by AOB (Hooper and Terry, 1979), suggesting that an active HAO in M. capsulatus Bath could also produce NO. We found that the steady-state level of haoA mRNA was lower at the 3′ end of the transcript than at the 5′ end, which may result in translation of a truncated HaoA (Hao') protein that lacks the tyrosine (Y467) needed for the intersubunit crosslink to catalytic heme 4 and a hydrophobic transmembrane spanning domain for anchoring HAO to the plasma membrane (Figure 2). Recently, the ability to form this crosslink was described as the critical acquisition during the evolution of HAO from an N-oxide-reducing octaheme cytochrome c nitrite reductase to an N-oxide oxidase (for a general discussion of HAO evolution, see (Klotz et al., 2008; Klotz and Stein, 2008)). The potential involvement of a C-terminal truncated Hao in the process of NO2 reduction to N-oxides is of particular interest as an additional mechanism by which M. capsulatus Bath may produce NO.

Figure 2
figure 2

Flow chart showing the proposed consequences of NH3-responsive differential transcription of the haoAB operon in Methylococcus capsulatus Bath. Question marks indicate yet unknown protein inventory; flag indicates promoter site; diamond, leaky terminator and full circle indicates terminator.

Implications of haoAB expression

The presence of an NH3-inducible haoA transcript encoding Hao suggests that M. capsulatus Bath could express a functional trimeric HAO complex. The possibility that M. capsulatus Bath possesses a functional HAO is supported by the previous work documenting the conservation of ligand-binding residues, the position of heme-binding motifs and an interprotein subunit tyrosine crosslink needed to establish the trimeric structure of the functional HAO complex, all of which aligned well with the structural features of HAO from AOB (Bergmann et al., 2005; Klotz et al., 2008). Although other laboratories have observed NO2 production by M. capsulatus Bath in the absence of NH3, the CH4+NH3 cultures in this study produced approximately 10 μM NO2, whereas CH4-only cultures produced no detectable NO2. These facts combined with our observation that haoA transcription was enhanced in the presence of NH3 suggest that M. capsulatus Bath has an active HAO. Additionally, cytochrome P460 expression did not respond to NH3 in this or previous studies (Bergmann et al., 1998, 2000), which further suggests that HAO could be the primary enzyme for the oxidation of NH2OH produced by pMMO or soluble MMO in M. capsulatus Bath.

The NH3 responsiveness of both cytS and haoAB transcription suggests that these genes could belong to a common regulatory network for the expression of NH3-inducible genes. Both genes are preceded by putative σ70 promoters (cytS, -35: TTGACg and -10: TATgAT; haoAB, -35: TTGACg and -10 TActtT), which indicates that the observed regulation of cytS and haoAB transcription by NH3 is likely not mediated by alternate sigma factors (Wosten, 1998). Instead, regulation by either small RNAs (Wassarman, 2002) or trans-acting DNA-binding proteins (Babu and Teichmann, 2003) could be involved. While candidate regulatory proteins have not yet been identified, the genome of M. capsulatus Bath encodes over 50 transcriptional regulators, many belonging to two component systems (Ward et al., 2004).

Potential physiological role of HAO in M. capsulatus Bath

In AOB, HAO is a key enzyme in the dissimilatory pathway that extracts electrons from NH3 and couples them to the generation of proton motive force. Hence, if M. capsulatus Bath possesses an active HAO, then it begs the question of whether this methanotroph can acquire metabolically useful electrons from NH3; the present literature lacks evidence that it can. The respiratory chain components of M. capsulatus Bath and AOB bear only partial resemblance: both contain cytochrome bc1 complex III and cytochrome aa3 terminal oxidase complex IV, both of which contribute to generation of proton motive force (DiSpirito et al., 2005; Hooper et al., 2005). On the other hand, M. capsulatus Bath lacks homologs of cytochromes c554 and cM552 of AOB, which in AOB are proposed to relay electrons from HAO to the quinone pool and hence contribute to proton motive force and usable reductant (Hooper et al., 2005; Klotz et al., 2008; and references therein). Methanotrophs use pMMO to convert CH4 to methanol, which is then oxidized to formaldehyde by a periplasmic methanol dehydrogenase. The extracted electrons are then relayed through two cytochromes (c555 and c553) to complex IV where they contribute to proton motive force and the reduction of oxygen (DiSpirito et al., 2005). Further, a dye-linked formaldehyde dehydrogenase donates a pair of electrons gained from the oxidation of formaldehyde in the cytoplasm directly to the cytochrome bc1 complex III, which contributes to proton motive force and serves pMMO with reductant needed for CH4 oxidation (DiSpirito et al., 2005).

If M. capsulatus Bath possesses an active HAO, it must have an as yet unidentified electron acceptor as a redox partner to continuously turn over substrate, and safely relay the extracted electrons to a terminal electron acceptor. If electrons extracted from NH2OH through oxidation by HAO were funneled into the quinone pool upstream of complex III, as in the case of the membrane-bound hydrogenase of M. capsulatus Bath (Hanczar et al., 2002) or relayed directly to cytochrome bc1, then NH2OH oxidation could contribute to the formation of proton motive force and/or serve pMMO with needed reductant, thereby supplementing the energy store of the cell. The M. capsulatus Bath genome encodes over 20 cytochromes c genes with unknown functions (Bergmann et al., 1999; Ward et al., 2004), offering a pool of potential redox partners to interact with HAO and link the oxidation and detoxification of NH2OH to the generation of metabolically useful electrons. Comparisons of the M. capsulatus Bath genome with genomes from lithotrophs that employ cytochromes c in their electron relays have identified the putative monoheme cytochrome c AAU915124 (COG4654) and the putative diheme cytochromes c AAU91311 (COG2863) and AAU91545 (COG2863) as candidates for a redox link to complexes III or IV (Klotz et al., 2006; Scott et al., 2006; Sievert et al., 2008). The gene encoding AAU91545 is clustered with cytS (AAU91546) as well as genes encoding a complex IV in the A-family of heme copper oxidases (AAU91541-AUU91544), offering candidates that might redox partner with either electron sink.


Our results suggest that M. capsulatus Bath possesses an active, NH3-responsive HAO. As the cytochrome P460 gene was expressed only at low levels regardless of whether ammonia was present, HAO appears to be a better candidate for the primary enzyme that oxidizes NH2OH to NO2. A more complete understanding of the role of methanotroph metabolism in the N-cycle now requires further biochemical confirmation, as well as identification of a suitable redox partner for HAO and pathway for final disposal of NH3-derived electrons in M. capsulatus Bath. Also of particular interest is the potential involvement of a truncated Hao in a process in which NO2 is immediately reduced to N-oxides. Ongoing studies of M. capsulatus Bath genes involved in nitrification and processing of N-oxides and their regulation will identify pathways used by methanotrophs in the global nitrogen cycle, and define the roles of reduced inorganic N-species in the physiology and growth of methanotrophs.

Experimental procedures

Culture conditions

M. capsulatus strain Bath was grown at 45 °C in nitrate mineral salts medium containing 10 μM CuSO4 (Whittenbury et al., 1970) under an initial headspace mixing ratio of 47:3:50 (CH4 to CO2 to air).

For RNA extraction, two independent cultures of M. capsulatus Bath in early exponential phase were harvested by centrifugation (6000 × g, 10 min, 25 °C) and washed twice with phosphate buffer (5.4 g Na2HPO4. 7H2O and 2.6 g KH2PO4 per liter distilled H2O). Pellets were resuspended in 1 ml of phosphate buffer and distributed to 0.2 l of nitrate mineral salts medium only (CH4-only treatment) or nitrate mineral salt medium supplemented with 5 mM (NH4)2SO4 (CH4+NH3 treatment). Cultures were incubated for 4 h under the above-described growth conditions. Following the incubation, cells were harvested for RNA extraction. The amount of NO2 in the medium was measured by a standard colorimetric method (Nicholas and Nason, 1957).

RNA extraction and cDNA synthesis

RNA was extracted using the FastRNA Pro Blue kit (Qbiogene, Irvine, CA, USA) according to the manufacturer's protocol. RNA pellets were resuspended in 100 μl of nuclease-free 0.1 mM EDTA (Ambion, Austin, TX, USA). Resuspended RNA was checked for integrity by visualization of ribosomal bands on an ethidium bromide-stained agarose gel and quantified by absorbance at 260 nm on a spectrophotometer (Beckman DU 640, Fullerton, CA, USA). RNA samples were treated with RQ1 RNase-free DNase (Promega, Madison, WI, USA) according to the manufacturer's recommended protocol, ethanol precipitated and resuspended in 20 μl of 0.1 mM EDTA and then reexamined for degradation as described above. A portion of the RNA was used in cDNA synthesis and the remainder was stored at −20 °C and used later as the non-reverse-transcribed control in qPCR.

RNA was converted to first-strand cDNA using 200 ng of random nonamer primer with Superscript III (Invitrogen, Carlsbad, CA, USA) reverse transcriptase according to the manufacturer's recommended protocol at an extension temperature of 55 °C for 60 min. The 20 μl reaction was diluted 1:5 with sterile H2O and stored at −20 °C for later analysis by qPCR.

Quantitative PCR

Primer sets (Table 1; Figure 1a) were designed with Primer3 software (; Rozen and Skaletsky, 2000) to target the following genes: 16S rRNA (rrsA; MCA_Mc16SA), pMMO subunit B (pmoB; MCA1796), hydroxylamine oxidoreductase (hao; MCA0956), orf2 (MCA0955), cytochrome P460 (cytL; MCA0524), cytochrome c′-β (cytS; MCA2394), NO reductase subunit C (norC; MCA2401) and NO reductase subunit B (norB; MCA2400). Commercially manufactured primers were resuspended to 5 μM and tested on M. capsulatus Bath genomic DNA (gDNA) for specificity and efficiency before use with cDNA.

Quantitative PCR of cDNA was performed in 20 μl reactions prepared in triplicate for each primer set using DyNAmo SYBR Green qPCR (New England Biolabs, Beverly, MA, USA) master mix according to the manufacturer's protocol. qPCR was also performed on nonreverse-transcribed RNA to check for carryover of gDNA. The fluorescence of the accumulated product was measured on a MJ Research thermal cycler (Bio-Rad Laboratories, Hercules, CA, USA) at each extension step using the following PCR parameters: initial denaturation at 94 °C for 5 min, 40–50 cycles consisting of 96 °C for 10 s, 55 °C for 10 s and 72 °C for 15 s. A melting curve was generated over a temperature range from 70 °C to 92 °C following PCR.

After log linearization of fluorescence versus cycle number curves and background subtraction of fluorescence, Ct values were obtained at a specific fluorescence level and used to compare the relative expression of haoA, orf2, cytL, cytS, norC and norB between CH4-only and NH3 treatments with rrsA and pmoB as reference genes. Data analysis was performed with the Relative Expression Software Tool (REST v.2005) (Pfaffl et al., 2002). Expression data are reported as expression ratios that represent the n-fold increase in the levels of the gene transcript in the NH3 treatment compared to the CH4-only treatment. Transcription of a gene was considered higher in the NH3 treatment if data analysis yielded an expression ratio >2 and a significant P-value (<0.05) for the randomization test performed by the REST program, which indicates the probability that the expression ratio obtained was due to a treatment effect, not chance.