Differential Regulation of the Three Methanol Methyltransferase Isozymes in Methanosarcina acetivorans C2A

Abstract

Genetic analysis of the three methanol-specific methyltransferase 1 operons (mtaCB1, mtaCB2, and mtaCB3) in Methanosarcina acetivorans led to the suggestion that each of them has a discrete function during growth on methanol, which might be reflected in differential gene regulation (Pritchett and Metcalf, Mol. Microbiol. 56:1183-1194, 2005). To test this suggestion, reporter gene fusions were constructed for each of the three operons, and their expression was examined under various growth conditions. Expression of the mtaCB1 and mtaCB2 fusions was 100-fold and 575-fold higher, respectively, in methanol-grown cells than in trimethylamine (TMA)-grown cells. The mtaCB3 fusion was expressed at low levels on methanol, TMA, and dimethylamine but was significantly upregulated on monomethylamine and acetate. When TMA- or acetate-grown cultures were shifted to methanol, the mtaCB1 fusion was expressed most highly during exponential phase, whereas the mtaCB2 fusion, although strongly induced prior to mtaCB1 expression, did not reach full expression levels until stationary phase. The mtaCB3 fusion was transiently expressed prior to entry into exponential phase during a TMA-to-methanol substrate shift experiment. When acetate-grown cells were shifted to medium containing both TMA and methanol, TMA utilization commenced prior to utilization of methanol; however, these two substrates were consumed simultaneously later in growth. Under these conditions expression of the mtaCB2 and mtaCB3 fusions was delayed, suggesting that methylamines may repress their expression.

Methane-producing archaea (methanoarchaea) are responsible for essentially all biological methane production on earth, yet most of these organisms are metabolically limited and are able to use one or, at most, two growth substrates. Moreover, all methanoarchaea are obligate methanogens. In this group, the Methanosarcina species are the only organisms that possess significant metabolic versatility. Many Methanosarcina species can use numerous methanogenic substrates, including H₂-CO₂, carbon monoxide, acetate, methanol, trimethylamine (TMA), dimethylamine (DMA), monomethylamine (MMA), methylsulfide, and dimethylsulfide (33). This metabolic diversity is reflected in the available genome sequences of three Methanosarcina species, which on average are more than twice the size of other known methanoarchaeal genomes. Interestingly, multiple copies of many of the genes that are specifically required for the use of alternate methanogenic substrates are present, and the multiple copies are conserved across all three Methanosarcina genomes that have been sequenced. For example, two or three copies of each of the genes required for entry of one-carbon compounds (C-1 compounds), such as methanol, TMA, DMA, and MMA, into the central methanogenic pathway are present. To date, little is known regarding the evolutionary advantage of having multiple isozymes of the proteins encoded by these genes; however, recent studies have begun to address this issue with respect to methanol utilization.

C-1 compounds are disproportionated to methane and carbon dioxide in a 3:1 ratio by using the reducing equivalents generated from oxidation of one methyl group to CO₂ to reduce three additional methyl groups to methane. These C-1 compounds enter the methanogenic pathway via activation by sets of substrate-specific methyltransferases designated methyltransferase 1 (MT1) and MT2 (Fig. 1). The methanol-specific MT1 consists of two protein components present at a 1:1 ratio (24, 30): MtaC, a 24-kDa corrinoid protein, and MtaB, a 49-kDa methyltransferase that transfers the methyl group from methanol to the corrinoid prosthetic group of MtaC. Subsequently, the MT2 reaction, catalyzed by the 38-kDa MtaA protein, transfers the methyl group from the corrinoid prosthetic group of MtaC to coenzyme M (mercaptoethanesulfonic acid) (28). The mtaC and mtaB genes comprise an operon, while mtaA is transcribed monocistronically (10, 24).

FIG. 1.

Simplified scheme for methanogenesis from methanol. Methanol (CH₃OH) is converted to methyl coenzyme M (2-mercaptoethanesulfonic acid) (CoM) by the concerted effort of three proteins. Methyltransferase 1 (MT1) is comprised of methanol:5-hydroxy-benzimidazolyl-cobamide methyltransferase, MtaB, and the corrinoid-binding protein, MtaC. The product of the MT1 reaction, methyl-MtaC is the substrate for a second methyltransferase (MT2 or MtaA), which transfers the methyl group from the corrinoid protein to coenzyme M. Methyl coenzyme M (CH₃-CoM) is then disproportionated, with one molecule oxidized to CO₂ to provide the electrons required for the reduction of three CH₃-CoM molecules to methane (not all steps are shown).

Analysis of the Methanosarcina acetivorans C2A genome sequence revealed the presence of three mtaCB operons (mtaCB1, mtaCB2, and mtaCB3) and two mtaA genes (mtaA1 and mtaA2) (9). Interestingly, these multiple gene copies are conserved in both Methanosarcina mazei Gö1 (5) and Methanosarcina barkeri strain Fusaro (GenBank accession no. NC007355). Furthermore, the multiplicity of methyltransferase-encoding genes is not limited to methanol-specific genes but is also seen for genes encoding enzymes specific for other C-1 substrates (7, 8, 26, 27). Thus, the advantages of multiple isozymes, whatever they are, may well apply to activation of all C-1 substrates.

Genetic analysis was recently used to address the role of the three mtaCB operons in M. acetivorans (21). A series of strains lacking the mtaCB1, mtaCB2, and mtaCB3 operons in all possible combinations was constructed. Strains with any two of the three operons deleted were able to grow on methanol, whereas strains with all three operons deleted were not able to grow on methanol, proving that mtaCB1, mtaCB2, and mtaCB3 all encode bona fide methanol-activating MT1 enzymes (however, this does not rule out the possibility that they can activate other substrates as well). Nevertheless, biochemical characterization of mutants showed that the three MT1 operons are not equivalent. Strains carrying only mtaCB1 had methyltransferase activity (measured during exponential phase during growth on methanol) similar to that of the wild type, whereas the methyltransferase activities of strains carrying only mtaCB2 or mtaCB3 were two- and fourfold lower, respectively. Interestingly, the presence of the mtaCB2 and mtaCB3 operons in addition to the mtaCB1 operon did not increase the overall methyltransferase activity. Thus, the function of the multiple gene copies cannot be simply to increase the overall methyltransferase activity. Although the growth rates and yields of most of these mutants were not affected, deletion of any one of the three operons resulted in prolonged lag phases (relative to the wild type) when the mutants were switched from other substrates to methanol. This effect was magnified in strains lacking two of the three operons. Strains carrying only mtaCB3 were particularly affected and exhibited much slower growth, very long lag phases, and reduced cell yields on methanol medium. Taken together, these data strongly suggest that the three isozymes play discrete roles during adaptation to and growth on methanol. If this is true, it is highly likely that the mtaCB1, mtaCB2, and mtaCB3 operons display differential gene regulation consistent with the discrete functions.

Recent data for Methanosarcina thermophila, M. mazei, and M. acetivorans support the idea that the MT1 isozymes are differentially regulated; however, these data are somewhat contradictory. Using two-dimensional gel electrophoresis of M. thermophila cell extracts coupled with mass spectrometric identification of proteins, Ding et al. showed that MtaC1, MtaB1, MtaC2, and MtaB2 were synthesized in methanol-grown cells but not in acetate-grown cells. Similar levels of MtaC3 and MtaB3 were found in extracts of both acetate- and methanol-grown cells (6). Similar results were obtained for M. acetivorans; MtaC1 and MtaB1 were ∼15-fold more abundant in methanol-grown cells, and MtaB2 was ∼33-fold more abundant in methanol-grown cells. However, in this organism the MtaC3 and MtaB3 proteins were reported to be induced approximately fivefold by growth on acetate (14, 15). These data led to the suggestion that expression of the mtaCB3 operon prior to exposure to methanol might allow rapid adaptation from acetate to methanol. Substantially different results were obtained in a recent DNA microarray study of M. mazei GÖ1(11). In this study, mtaCB1 expression was induced ∼30-fold during growth on methanol compared to growth on acetate. However, the mtaCB2 operon was not induced by methanol but was induced ∼10-fold by growth on acetate. Whether these data reflect true differences between the species or whether they result from inherent limitations of the experimental methods remains to be seen.

The experimental complications involved in examining highly homologous genes led us to directly examine the regulation of the three mtaCB operons using reporter gene fusions to the promoters of each operon. This method, made possible by recent advances in the genetic analysis of Methanosarcina species, unambiguously discriminates between the three gene copies and is much simpler than other methods, and it allowed us to examine gene expression over a broader range of growth conditions. Here we describe a comprehensive analysis of mtaCB gene regulation in M. acetivorans. In addition to examining the expression of the operons during exponential growth on methanol and acetate, we examined the levels of expression on TMA, DMA, and MMA, the temporal pattern of expression throughout growth during shifts from methylamines to methanol, from acetate to methanol, and from acetate to trimethylamine, and the regulation of expression in the presence of multiple substrates. Our data provide new insight into the roles of multiple gene copies in methanoarchaea.

MATERIALS AND METHODS

Bacterial strains, media, and growth conditions.

Escherichia coli cells were grown under standard conditions (31). Methanosarcina strains were grown with single-cell morphology (25) at 37°C in high-salt (HS) broth containing 125 mM methanol, 50 mM TMA, 50 mM DMA, 50 mM MMA, or 120 mM acetate, as appropriate (18). Growth on media solidified with 1.5% agar was as described by Boccazzi et al. (3). All plating manipulations were carried out under strictly anaerobic conditions in an anaerobic glove box. Plates containing solid media were incubated in an intrachamber anaerobic incubator as described previously (19). Puromycin (CalBiochem, San Diego, CA) was added from sterile, anaerobic stocks at a final concentration of 2 μg/ml for selection of Methanosarcina strains carrying puromycin transacetylase (pac). The purine analog 8-aza-2,6-diaminopurine (Sigma, St. Louis, MO) was added from sterile, anaerobic stocks at a final concentration of 20 μg/ml for selection against the hpt gene.

Methanosarcina strains used in the study were constructed by markerless gene exchange as described previously (22). All of the strains used are derivatives of M. acetivorans DSM 2834. WWM12 (hpt::PmtaCB1::uidA) was made using pMP58, WWM11 (hpt::PmtaCB2::uidA) was made using pMP59, WWM62 (hpt::PmtaCB3::uidA) was made using pMR52, and WWM63 (hpt::PmcrB::uidA) was made using pMR53. The plasmids used and their construction are summarized in Table 1.

TABLE 1.

Plasmids used in this study

Plasmid	Description and/or construction	Reference
pWM356	pBluescript KS with the BamHI site filled	This study
pMP42	Vector for insertion into hpt locus	22
pJK41	Apr Pmr cloning vector, oriR6K replicons	16
pMP45	NdeI/BglII-cut uidA PCR product [obtained using primers uidA(NdeI) and uidA(Bgl2)] cloned into BglII/BamHI-cut pWM368	22
pMP30	SpeI/MluI-digested up-mtaC1 PCR product (obtained using primers SpeI2mtaC1 and MluINdeICI) and MluI/NotI-digested dn-mtaC1 PCR product (obtained using primers MluIBamHB1 and NotI2mtaB1) cloned into SpeI/NotI-digested pJK41	21
pMP31	SpeI/MluI-digested up-mtaC1 PCR product (obtained using primers SpeI2mtaC2 and MluINdeIC2) and MluI/NotI-digested dn-mtaC1 PCR product (obtained using primers MluIBamHB2 and NotI2mtaB2) cloned into SpeI/NotI-digested pJK41	21
pMP38	SpeI/NotI-digested ΔCB1 fragment of pMP30 cloned into SpeI/NotI-digested pBluescript (-BamHI)	This study
pMP39	SpeI/NotI-digested ΔCB2 fragment of pMP30 cloned into SpeI/NotI-digested pBluescript (-BamHI)	This study
pMP51	NdeI/BglII uidA fragment from pMP45 cloned into NdeI/BamHI-digested pMP38	This study
pMP52	NdeI/BglII uidA fragment from pMP45 cloned into NdeI/BamHI-digested pMP39	This study
pMP58	SpeI/NotI fragment of pMP51 carrying PmtaCB1-uidA cloned into AvrII/NotI fragment of pMP42	23
pMP59	SpeI/NotI fragment of pMP51 carrying PmtaCB2-uidA cloned into AvrII/NotI fragment of pMP42	This study
pAMG46	Derivative of pMP42 with different cloning sites	23
pMR51	NotI/NdeI fragment of pMP58 cloned into NotI/NdeI-digested pAMG46	This study
pMR52	NheI/NdeI-digested PmtaCB3 PCR product (obtained using primers oCB3/5′ and oCB3/3′) cloned into NheI/NdeI-digested pMR51	This study
pMR53	NheI/NdeI-digested PmcrB PCR product [obtained using primers oPmcrB(ac)/5′ and oPmcrB(ac)/3′] cloned into NheI/NdeI-digested pMR51	This study
pWM368	Methanosarcina barkeri Fusaro mcrB promoter source	32

Growth conditions for reporter gene assays.

For the single-time reporter gene assays, the strains were grown on substrates for at least 25 generations (at least five transfers using 1% inocula) and then harvested at mid-exponential phase (optical density at 600 nm [OD₆₀₀] for TMA or methanol, 0.4 to 0.5; OD₆₀₀ for DMA, 0.3; OD₆₀₀ for acetate and MMA, 0.1 to 0.2) for β-glucuronidase assays. For the substrate shift experiments cultures were grown for at least 25 generations on TMA or acetate, harvested at the mid-exponential phase of growth by centrifugation, washed two times with plain HS medium to remove residual substrate, and then resuspended in HS medium (500 ml) with the appropriate substrate(s) to an initial OD₄₂₀ of 0.1. At various times, samples were withdrawn (30-ml samples during the lag phase of growth and 10-ml samples during the exponential phase of growth for acetate substrate shift experiments; 20-ml samples during the lag phase of growth and 10-ml samples during the exponential phase of growth for TMA substrate shift experiments) to measure growth (OD₄₂₀), the amount of methane produced, the methanol, TMA, DMA, and MMA concentrations, and the β-glucuronidase activity.

DNA methods.

Standard methods were used throughout this study for isolation and manipulation of plasmid DNA from E. coli (1). The plasmids and primers used in this study are shown in Tables 1 and 2. Genomic DNA from M. acetivorans was isolated as described previously (22). DNA hybridization was performed using the DIG system (Roche, Mannheim, Germany). Magnagraph nylon transfer membranes were obtained from Osmonics (Westborough, MA). DNA sequences were determined using double-stranded templates by the W. M. Keck Center for Comparative and Functional Genomics, University of Illinois.

TABLE 2.

Primers used in this study

Primer	Sequencea	Added site(s)
SpeI2mtaC1	CCGCCGACTAGTTTATCTTTTTCAACCAGGGTAAGG	SpeI
MluINdeICI	CCGCCGACGCGTCATATGTCTAAACCTCCATTTAG	MluI, NdeI
MluIBamHB1	CCGCCGACGCGTGGATCCGCCCTCAGTTCTCTTTTTC	MluI, BamHI
NotI2mtaB1	CCGCCGGCGGCCGCCAGGAGGGATGGAAAAAGG	NotI
SpeI2mtaC2	CCGCCGACTAGTTTCAGCCTTCCAGAAAAACC	SpeI
MluINdeIC2	CCGCCGACGCGTCATATGTTAACCTCCATTTTAATAATGAAGC	MluI, NdeI
MluIBamHB2	CCGCCGACGCGTGGATCCAATTTTTTTCAAAAAATGGC	MluI, BamHI
NotI2mtaB2	CCGCCGGCGGCCGCGTCCTGCATACGAGATGTTCC	Not
uidA(NdeI)	GGGGGGCATATGTTACGTCCTGTAGAAACCC	NdeI
uidA(Bgl2)	GGGGGGAGATCTGATCATTAACGGCGCAGTACCG	BglII
oCB3/5′	GGAATTCCATATGCTAGCTAGCCGCGATAGATATTTGAAAAACATCTATC	NdeI, AceII
oCB3/3′	GAATTCCCATATGTTTAACCTCCATTTTAGTATTTGAGGAGTAAATATC	NdeI
oPmcrB(ac)/5′	GGAATTCCATATGCTAGCTAGCCAGAGGGTCTTTTCGAGGAC	NdeI, AceII
oPmcrB(ac)/3′	GAATTCCCATATGAATTTCCTCCTTAATTTATTAAAATCATTTTGGG	NdeI

^aAdded restriction sites are underlined.

Transformation methods.

E. coli strains were transformed by electroporation using an E. coli Gene Pulser (Bio-Rad, Hercules, CA) as recommended by the supplier. Liposome-mediated transformation was used for Methanosarcina as described previously (3, 17).

Extract preparation and β-glucuronidase assay.

The method used for preparation of cell extracts and the β-glucuronidase assay method were methods described previously (23).

Determination of metabolites.

Methanogenic substrates and products were quantified using a Hewlett-Packard gas chromatograph (5890 Series II) equipped with a flame ionization detector. A stainless steel 80/120 Carbopack B column [Sigma-Aldrich (Supelco), St. Louis, MO] with He as the carrier gas was used at a constant temperature of 120°C for determination of methane and methanol contents. A glass 60/80 Carbopack B column [Sigma-Aldrich (Supelco), St. Louis, MO] with He as the carrier gas was used at a constant temperature of 95°C for measurement of TMA, DMA, and MMA. For analysis of TMA, DMA, and MMA, the samples were diluted 1:2 in 1% KOH before injection into the gas chromatograph. The column was washed with 10 μl distilled water, followed by 10 μl 1% KOH, between runs.

RESULTS

Construction of PmtaCB1::uidA, PmtaCB2::uidA, and PmtaCB3::uidA gene fusions.

The uidA gene of E. coli encodes the easily assayable enzyme β-glucuronidase, which has previously been used as a reporter gene in Methanosarcina and other methanoarchaea (2, 12, 23). We constructed uidA reporter gene fusions to the promoter regions of the mtaCB1 (23), mtaCB2, and mtaCB3 operons. In these constructs the start codon of the uidA gene was superimposed with the start codon of the corresponding mtaC gene, and the constructs carried ca. 1 kb of upstream DNA. However, the mtaC3 gene apparently utilizes a TTG start codon (21), which was changed to ATG in the PmtaC3::uidA fusion to facilitate the cloning steps and to maintain uniformity with the other fusions. As a control, we utilized a fusion to the promoter of the mcrB gene of M. acetivorans C2A, which is the first gene of the mcrBDCGA operon. This operon encodes methyl-coenzyme M reductase, which catalyzes the terminal step of methanogenesis during growth on all methanogenic substrates. It should be noted that the promoters for these operons have not been precisely mapped. We assumed that all required regulatory sequences were present within the 1-kb region upstream of the coding region because all three mtaCB fusions were expressed on one or more substrates, although it is possible that additional control sequences were present, either further upstream or within the coding sequence. Each of the fusion constructs was subsequently integrated as a single copy into the hpt locus on the M. acetivorans C2A chromosome using the markerless exchange method (22). All strains were verified to have the correct insertions into the hpt locus by DNA hybridization (data not shown).

Expression of mtaCB operons on various methanogenic substrates.

As described above, our previous gene deletion experiments suggested discrete roles for each of the three mtaCB operons, which might be reflected in gene regulation. Thus, we examined the expression of the three mtaCB::uidA fusions at mid-exponential phase during growth in media containing either methanol, TMA, DMA, MMA, or acetate (Table 3). The mtaCB1 and mtaCB2 promoter gene fusions were highly upregulated on methanol compared to the expression with the other substrates tested. The mtaCB1 promoter gene fusion was induced ∼100-fold during growth on methanol (compared to expression on TMA), whereas the mtaCB2 promoter gene fusion was induced ∼575-fold. These data strongly support previous molecular, genetic, and biochemical data suggesting that both operons are specifically involved in the metabolism of methanol. Interestingly, the mtaCB1 promoter gene fusion was expressed only during growth on methanol, while expression of the mtaCB2 promoter gene fusion was also induced during growth on both acetate and MMA. Thus, despite the observation that both promoter gene fusions are expressed during growth on methanol, the two operons are clearly differentially regulated. In contrast, expression of the mtaCB3 promoter gene fusion was not induced in response to methanol at mid-exponential phase. Instead, this promoter gene fusion appeared to be expressed at a low, constitutive level during growth on methanol, TMA, or DMA. However, expression of the mtaCB3 promoter gene fusion increased ∼10-fold during growth on acetate and ∼5-fold during growth on MMA (relative to expression on TMA) (Table 3). Although the expression of the control mcrB promoter gene fusion also increased during growth on acetate, the level of induction (approximately threefold) was much less than the expression of the mtaCB3 promoter gene fusion. Furthermore, the control promoter gene fusion was not induced on MMA. Thus, the expression pattern of the mtaCB3 promoter gene fusion appears to suggest that there is substrate-specific gene regulation that is clearly different than that of the mtaCB1 and mtaCB2 promoter gene fusions. The increased expression of the mtaCB3 promoter gene fusion suggests that these genes might play an important role during growth on acetate, despite the observation that mtaCB3 mutants do not have any known phenotype on acetate (21).

TABLE 3.

β-Glucuronidase activities of uidA translational fusions to mtaCB1, mtaCB2, mtaCB3, and mcrB in cells grown on various methanogenic substrates^a

Fusion	β-Glucuronidase activities with the following substrates:
	Methanol	TMA	DMA	MMA	Acetate
mtaC1b	315 ± 47	3.3 ± 0.3	1.14 ± 0.2	1.1 ± 0.1	3.6 ± 0.7
mtaC2	1,666 ± 140	2.9 ± 0.8	1.9 ± 0.2	15 ± 2.7	128 ± 28
mtaC3	25 ± 6	36 ± 5	40 ± 3	161 ± 40	241 ± 19
mcrB	548 ± 53	802 ± 65	432 ± 55	605 ± 31	2,159 ± 179

^aActivity is expressed in mU/mg protein (nmol min⁻¹ mg protein⁻¹). The values are averages and standard errors for nine independent measurements.

^bData for this fusion were originally reported by Rother et al. (23) and are included for completeness.

Growth phase-dependent expression of mtaCB isozymes.

In an effort to further delineate the regulatory differences between mtaCB1, mtaCB2, and mtaCB3, the temporal expression pattern of each promoter gene fusion strain was analyzed when TMA-adapted cultures were switched to methanol (Fig. 2). In addition, substrate shift experiments were performed by shifting acetate-grown cultures to either methanol (Fig. 3) or TMA (data not shown). At each time, reporter gene activity, growth (OD₄₂₀), methanol consumption, and methane production (data not shown) were also measured.

FIG. 2.

PmtaC::uidA expression during a substrate shift from trimethylamine to methanol. Cultures adapted to TMA (50 mM) were switched to medium containing methanol (125 mM). The OD₄₂₀ indicates growth, and the methanol concentration was determined by gas chromatography. The β-glucuronidase activity indicates expression from the specific promoters. (A) PmtaC1::uidA; (B) PmtaC2::uidA; (C) PmtaC3::uidA. The y axes indicate log OD₄₂₀ (red), methanol concentration (dark blue), and β-glucuronidase activity (light blue). The β-glucuronidase activities of the three strains were very different, and thus the scales used for specific activity are not identical. Triplicate experiments were performed for each fusion strain. Equivalent results were obtained for all replicates; a representative curve is shown for each experiment.

FIG. 3.

PmtaC::uidA expression during a substrate shift from acetate to methanol. Cultures adapted to acetate (120 mM) were switched to medium containing methanol (125 mM). The OD₄₂₀ indicates growth, and the methanol concentration was determined by gas chromatography. The β-glucuronidase activity indicates expression from the specific promoters. (A) PmtaC1::uidA; (B) PmtaC2::uidA; (C) PmtaC3::uidA. The y axes indicate log OD₄₂₀ (red), methanol concentration (dark blue), and β-glucuronidase activity (light blue). The β-glucuronidase activities of the three strains were very different, and thus the scales used for specific activity are not identical. Duplicate experiments were performed for each fusion strain. Equivalent results were obtained for all replicates; a representative curve is shown for each experiment.

A significant lag occurred prior to expression of any of the reporter gene fusions. None of the mtaCB promoter gene fusions were expressed during most of this period (75 h for mtaCB1 and mtaCB3 and 50 h for the mtaCB2 promoter), which argues against a simple model of transcriptional regulation involving a methanol-sensing repressor protein. The mtaCB1 promoter gene fusion was expressed early in exponential growth, but only after methanol utilization and methane production (data not shown) had already begun. This indicates that another methanol-specific MT1 (either MtaCB2 or MtaCB3 or both) was responsible for the activation of methanol during the earliest phase of growth. Both the mtaCB2 and mtaCB3 promoter gene fusions were expressed during this preexponential phase of growth, before significant methanol utilization, methane production (data not shown), or cell growth was observed. Although induction of the mtaC3 promoter gene fusion was transient during the TMA-to-methanol substrate shift experiment, the level of expression was significant (at least threefold above the basal level of expression) and reproducible in all replicates of this experiment, suggesting a specific role for mtaCB3 during this substrate shift. The induction of the mtaCB3 promoter gene fusion was not clear during the acetate-to-methanol substrate shift experiment because this operon was induced by pregrowth on acetate. Expression of both the mtaCB1 and mtaCB2 promoter gene fusions peaked during early stationary phase.

During the acetate-to-TMA substrate shift experiment (data not shown) the mtaCB1 promoter gene fusion was not expressed at any time during growth. However, both the mtaCB2 and mtaCB3 promoter gene fusions were expressed later during growth. The expression of the mtaCB2 promoter gene fusion increased when DMA appeared in the medium, whereas the expression of the mtaCB3 promoter gene fusion increased when MMA appeared in the medium, which is consistent with the observations from the mid-exponential measurements of expression of these promoters (Table 3).

Expression of mtaCB operons in the presence of both TMA and methanol.

To examine the effect of the presence of multiple substrates on the expression of the three mtaCB promoter gene fusions, cultures were grown to mid-exponential phase in acetate medium and then shifted to a medium containing both TMA and methanol (Fig. 4). Reporter gene activity, growth, methane production (data not shown), and the concentrations of methanol, TMA, DMA, and MMA were monitored throughout the experiment. (DMA and MMA are the products of TMA utilization and are themselves substrates for growth and methanogenesis.)

FIG. 4.

PmtaC::uidA expression during a substrate shift from acetate to methanol and trimethylamine. Cultures adapted to acetate (120 mM) were switched to medium containing both methanol (125 mM) and TMA (50 mM). The OD₄₂₀ indicates growth, and the methanol, TMA, DMA, and MMA concentrations were determined by gas chromatography. The β-glucuronidase specific activity indicates expression from the specific promoters. (A) PmtaC1::uidA; (B) PmtaC2::uidA; (C) PmtaC3::uidA. The y axes indicate log OD₄₂₀ (red), methanol concentration (dark blue), and β-glucuronidase activity. The β-glucuronidase activities of the strains were very different, and thus the scales used for specific activity are not the same. Duplicate experiments were performed for each fusion strain. Equivalent results were obtained for both replicates; a representative curve is shown for each experiment.

Two significant trends were observed in the substrate shift experiment in which there was a shift from acetate to a combination of TMA and methanol (Fig. 4). First, we observed that the cells used TMA almost exclusively for initial growth before any detectable decrease in the methanol concentration. The early accumulation of DMA in the medium supports the hypothesis that there was early utilization of TMA for growth. Both DMA and MMA started accumulating before rapid utilization of methanol commenced. Finally, the growth curve displays a diauxic pattern (which is more pronounced when the data are plotted on a linear scale), which correlates with the nearly simultaneous depletion of TMA, DMA, and MMA. In keeping with the idea that TMA is preferred over methanol for initial growth, there was no observable induction of any of the mtaCB promoter gene fusions during the initial stages of growth. (Note that the expression of mtaCB2 and mtaCB3 promoter gene fusions, which was induced by pregrowth on acetate, actually declined during the lag phase prior to initiation of growth.) Second, the relative order and degree of expression of the mtaCB promoter gene fusions were substantially altered during the substrate shift experiment in which there was a shift from acetate to a combination of methanol and methylamines. As described above, both the mtaCB2 and mtaCB3 promoter gene fusions were expressed prior to the expression of the mtaCB1 promoter gene fusion when cells were shifted from either TMA or acetate to methanol (Fig. 2 and 3). In contrast, when cells were switched from acetate to the mixture of TMA and methanol, the mtaCB1 promoter gene fusion was expressed well before the mtaCB2 promoter gene fusion was expressed. Furthermore, the level of expression of the mtaCB1 promoter gene fusion was approximately threefold higher during the switch from acetate to TMA plus methanol than during the switch from TMA to methanol. It is also notable that neither the mtaCB2 promoter gene fusion nor the mtaCB3 promoter gene fusion was induced until ca. 90% of the TMA had been utilized. Moreover, the level of expression of the mtaCB2 promoter gene fusion with the combination of TMA and methanol was only 5% of the expression level observed after the switch from TMA to methanol. (It should be noted that in the TMA-to-methanol substrate shift experiment the cells were washed twice prior to transfer to remove residual methylamines, ensuring that their presence did not affect expression of the three mtaCB promoter gene fusions.) In addition, the induction of the mtaCB3 promoter gene fusion was concurrent with the accumulation of MMA in the medium, which was in accordance with the earlier observation that the expression of the mtaCB3 promoter gene fusion was induced on MMA.

DISCUSSION

The data presented here clearly show that the mtaCB1, mtaCB2, and mtaCB3 promoter gene fusions are differentially regulated in M. acetivorans, strongly supporting the idea that the three MT1 methyltransferases play discrete roles in the cell. In the present study we showed that mtaCB1 is expressed only during exponential growth on methanol. This is consistent with the purification of MtaCB1 as the sole MT1 purified from methanol-grown M. barkeri (24). In addition, our previous observation shows that MtaCB1 has the highest methyltransferase activity (21). Based on these data, we suggest that MtaCB1 is the primary MT1 utilized during exponential growth on methanol. The roles of MtaCB2 and MtaCB3, however, are somewhat more difficult to define. It has been proposed that prior synthesis of MtaCB3 during growth on acetate allows efficient switching from acetate to methanol (6, 15). Our finding that mtaCB3 is expressed on all substrates is consistent with this idea. Moreover, the transient induction of this operon right at the point of switching to methanol, but not at the point of switching to TMA from acetate (Fig. 2), supports a specific role in substrate switching to methanol. However, it should be remembered that mutants lacking mtaCB3 have shorter lag times when they are switched from TMA to methanol than mutants lacking either mtaCB1 or mtaCB2 have (21). Thus, mtaCB3 is not absolutely required for switching, nor does MtaCB3 appear to be the only enzyme involved. We suggest that mtaCB2 probably also plays a significant role in substrate switching. This idea is consistent with the long lag phase during switching to methanol that is observed in mtaCB2 mutants (21) and with the observation made here that, in the absence of substrates other than methanol, mtaCB2 is the first of the three operons to be expressed. However, the highest level of mtaCB2 and mtaCB1 expression is observed as the cells enter stationary phase. This suggests the possibility that the cells recognize methanol depletion and accumulate high levels of MtaCB2 and MtaCB1 in order to be prepared for rapid growth onset should the substrate become available again. An alternative, and equally consistent, model involves MtaCB2 as a high-affinity (low-K_m) methanol-activating enzyme. Thus, as the concentration of methanol falls during growth in batch culture, the need for a high-affinity enzyme and correspondingly the expression of mtaCB2 increase, reaching the maximum level as the cells enter stationary phase. In addition, if induction of a methanol transporter were required, the operon would be expressed early in growth. Thus, prior to expression of this putative transporter, cytoplasmic methanol concentrations would be low, requiring the low-K_m enzyme, explaining the prolonged lag phase of mtaCB2 mutants. Because of the trade-off between high affinity and enzyme velocity (low V_max), this would also explain the low methyltransferase activity of strains expressing only MtaCB2 (21), which is especially puzzling given the very high levels of expression of the mtaCB2 reporter gene fusion (almost 10-fold higher than the levels of expression of mtaCB1 [Table 3]).

Our data also clearly show that regulation of mtaCB operons is more complex than a simple response to methanol and growth phase. In the substrate shift experiment in which there was a shift from acetate to methanol plus TMA, expression of mtaCB2 and mtaCB3 did not occur until ca. 90% of the methylamine was consumed. Therefore, it appears that methylamines might inhibit expression of mtaCB2 and mtaCB3 operons, although to different degrees. It is also interesting that both the mtaCB2 and mtaCB3 operons are specifically induced during growth on MMA (Table 3 and Fig. 4) and are induced even more on acetate (Table 3). Both methanol and TMA have been shown to be important substrates for methanogens in marine sediments (20). Thus, in order to understand the preferential utilization of TMA over methanol for initial growth, we turned to the energetics of methanogenesis from various methanogenic substrates. It has been argued that the low energy yield available from methanogenesis from acetate requires the prior synthesis of the enzymes needed for use of a new substrate in order to promote efficient switching (6). We believe that the same argument would hold for methanogenesis from MMA. However, this must be examined carefully. Typically, the energetics of methanogenesis are reported as ΔG°′/mol CH₄ (Table 4). On this basis, acetate is clearly the poorest substrate, TMA, DMA, and MMA are about equal, and methanol is the best substrate. This is a useful measure, in that it reveals the free energy available from electron transport on a per-electron basis. Moreover, the values are consistent with the observation that the highest growth rates are obtained during growth on methanol (21). Nevertheless, when substrate preference is considered, it is probably much more relevant to compare the ΔG°′/mol substrate. On this basis TMA is the best substrate by a wide margin, followed by DMA, methanol, MMA, and finally acetate. With this in mind, our data are consistent with hierarchical regulation of the MT1-encoding genes based on the quality of the available substrates, analogous to catabolite repression in enteric bacteria. This model readily explains the preference for TMA over methanol in the substrate shift experiment, the lack of expression of the mtaCB operons in the presence of TMA, and the induction of the putative switching enzymes (MtaCB3 and MtaCB2) during growth on low-energy substrates.

TABLE 4.

Free energy of methanogenic reactions^a

Substrate	Reaction	ΔG°′ (kJ)
		Per reaction	Per mol CH4	Per mol substrate
Methanol	4CH3OH → 3CH4 + CO2 + 2H2O	−318	−106	−80
TMA	4(CH3)3N + 6H2O → 9CH4 + 3CO2 + 4NH3	−682	−76	−171
DMA	2(CH3)2NH + 2H2O → 3CH4 + CO2 + 2NH3	−224	−75	−112
MMA	4CH3NH2 + 2H2O → 3CH4 + CO2 + 4NH3	−230	−77	−57
Acetate	CH3COO− + H+ → CH4 + CO2	−36	−36	−36

^aValues were calculated from data of Thauer et al. (29).

At present, it is unclear how the complex differential regulation of the mtaCB operons is achieved, but it seems clear that multiple types of information must be integrated. Our data eliminate many of the simple regulatory paradigms. For example, a methanol-sensing activator or repressor protein alone does not explain the low level of expression of the operons when both methanol and TMA are present. Nor does a TMA-sensing repressor explain the specific induction of mtaCB3 and mtaCB2 on MMA and acetate. Further examination of this process is clearly warranted, and such studies are being performed in our laboratory. In this regard, the mtaCB operons may be among the most fertile candidates for study of gene regulation in archaea. To our knowledge, the level of regulation measured in our mtaCB1 and mtaCB2 fusions is unprecedented in the archaeal domain. We have recently shown that it is possible to select constitutive mutants by fusing the mtaC1 promoter to an essential gene (23), suggesting that genetic approaches should be fruitful in unraveling this mystery.

Finally, given the wealth of data coming from proteomic and microarray projects, one might question the need for single-promoter reporter gene studies. Indeed, our data are qualitatively consistent with the M. acetivorans proteomic data obtained by Li et al. (14, 15), although the fold regulation that we observed is much higher. Because the proteomic studies rely on quantification of stained protein spots on a two-dimensional gel, the values must be considered minimum estimates. Since our fusions maintain both transcriptional and translational control sequences, we believe that the reporter genes provide a more accurate reflection of the actual degree of mtaCB gene regulation. Our data are, however, at odds with the results obtained with M. mazei microarrays (11). Although it is possible that the data reflect real differences in gene regulation between the two species, we feel that it is likely that the difficulties in analysis of highly homologous genes may account for the discrepancies. Accordingly, it seems very likely that the multiple mtaCB operons cross-hybridize during nucleic acid hybridization experiments. This problem is well recognized and was acknowledged by the authors of the M. mazei study (4, 11, 13). The genetic approach used here has no difficulty in discriminating between the different gene copies. Moreover, the dynamic studies described here, although technically possible, would be extremely costly and time-consuming using either microarray or proteomic approaches. In contrast, the reporter gene approach used here is rapid, simple, and inexpensive. However, reporter gene fusions have a few disadvantages compared to microarray analysis. In particular, although gene fusions can readily be used to show activation of expression, they have limited capacity for demonstrating gene shutoff (due to the stability of the reporter protein). In addition, it is formally possible that introduction of an additional copy of the promoter under study may perturb the regulatory system. With these caveats in mind, our studies revealed differential and timely regulation of important genes in the methanol utilization pathway, showing the complex interaction between multiple substrates and growth phases and providing new insight into the roles of multiple copies of many methanogenesis genes in Methanosarcina.