Mechanism for Stabilizing mRNAs Involved in Methanol-Dependent Methanogenesis of Cold-Adaptive Methanosarcina mazei zm-15

Abstract

Methylotrophic methanogenesis predominates at low temperatures in the cold Zoige wetland in Tibet. To elucidate the basis of cold-adapted methanogenesis in these habitats, Methanosarcina mazei zm-15 was isolated, and the molecular basis of its cold activity was studied. For this strain, aceticlastic methanogenesis was reduced 7.7-fold during growth at 15°C versus 30°C. Methanol-derived methanogenesis decreased only 3-fold under the same conditions, suggesting that it is more cold adaptive. Reverse transcription-quantitative PCR (RT-qPCR) detected <2-fold difference in the transcript abundances of mtaA1, mtaB1, and mtaC1, the methanol methyltransferase (Mta) genes, in 30°C versus 15°C culture, while ackA and pta mRNAs, encoding acetate kinase (Ack) and phosphotransacetylase (Pta) in aceticlastic methanogenesis, were 4.5- and 6.8-fold higher in 30°C culture than in 15°C culture. The in vivo half-lives of mtaA1 and mtaC1B1 mRNAs were similar in 30°C and 15°C cultures. However, the pta-ackA mRNA half-life was significantly reduced in 15°C culture compared to 30°C culture. Using circularized RNA RT-PCR, large 5′ untranslated regions (UTRs) (270 nucleotides [nt] and 238 nt) were identified for mtaA1 and mtaC1B1 mRNAs, while only a 27-nt 5′ UTR was present in the pta-ackA transcript. Removal of the 5′ UTRs significantly reduced the in vitro half-lives of mtaA1 and mtaC1B1 mRNAs. Remarkably, fusion of the mtaA1 or mtaC1B1 5′ UTRs to pta-ackA mRNA increased its in vitro half-life at both 30°C and 15°C. These results demonstrate that the large 5′ UTRs significantly enhance the stability of the mRNAs involved in methanol-derived methanogenesis in the cold-adaptive M. mazei zm-15.

INTRODUCTION

Representatives of the order Methanosarcinales dominate the methanogenic community in wetlands located in cold regions (1, 2), where they comprise diverse physiological groups, including the versatile Methanosarcina spp., which use acetate, methyl amines, methanol, and H₂/CO₂ as substrates for methanogenesis, and the obligate methylotrophic (Methanococcoides and Methanolobus) and obligate aceticlastic (Methanosaeata) methanogens. Previously, we determined that most of the methane released from the cold Zoige wetland on the Tibetan plateau was derived from methanol or acetate, whereas methanol supported the highest rate of CH₄ formation in soil enrichments. The rate was even higher at 15°C than at 30°C (3), suggesting that methanol-derived methanogenesis by this community was most active in the cold.

Methylotrophic or aceticlastic methanogenesis requires that the precursors be converted to methyl-coenzyme M (CoM) prior to the reduction of methyl-CoM to CH₄. When methanol is the substrate, the methanol-coenzyme M methyltransferase complex catalyzes the conversion of methanol to methyl-CoM. This complex comprises three proteins: a methanol-specific methyltransferase, MtaB (methanol-corrinoid methyltransferase), for transferring the methyl to its cognate corrinoid protein;MtaC (methanol corrinoid protein); and methyltransferase 2 (MtaA; methylcobalamin-coenzyme M methyltransferase), which catalyzes the transfer of the methyl group from MtaC to CoM. In the sequenced methanosarcinal genomes, three copies of mtaC and mtaB and two copies of mtaA are found (4). In aceticlastic methanogenesis, acetate is first activated to acetyl-coenzyme A (CoA) by acetate kinase (Ack) and phosphotransacetylase (Pta). Acetyl-CoA is then cleaved into an enzyme-bound methyl group and CO₂ by acetyl-CoA synthase (ACS)/CO dehydrogenase (CODH). The methyl carbon is then transferred to CoM via the C₁ carrier tetrahydrosarcinapterin (5).

Opulencia et al. (6) indicated that the mtaA and mtaCB transcripts exhibited different stabilities, implying posttranscriptional regulation. mRNA stability is a major determinant of posttranscriptional control of gene expression (7, 8) and plays significant roles in cellular adaptation, because of its prompt response to environmental changes (9).

To investigate the impact of mRNA stability on cold-active methanol-derived methanogenesis, in this study, a psychrotolerant Methanosarcina mazei strain, zm-15, which performs both methylotrophic and aceticlastic methanogenesis, was isolated from the cold Zoige wetland in Tibet. We found that in this cold-adapted organism, methanol supported cold-active methanogenesis more than acetate, which was attributed, at least partially, to the longer life span of the mRNAs of the key enzymes.

MATERIALS AND METHODS

Soil sample collection.

Soil covered by Eleocharis valleculosa at a depth of 10 to 30 cm was collected from the Zoige wetland (33°56′N, 102°52′E; altitude, 3,430 to 3,460 m), located on the Tibetan Plateau, in April 2007. The soil samples were stored in sterile serum bottles sealed with butyl rubber stoppers (with N₂ as the gas phase) and kept in an ice-cold box during transportation to the laboratory.

DNA extraction, 16S rRNA sequencing, and phylogenetic analysis.

Total DNA was extracted from the soil samples (approximately 5 g) and purified with a FastDNA Spin kit for Soil (MP Biomedicals, Solon, OH, USA). The purified DNA was stored at −20°C.

For PCR amplification of methanogenic 16S rRNA genes, the methanogen-specific primers Met83F and Met1340R (see Table S1 in the supplemental material) were used (10) with Taq DNA polymerase (TaKaRa, Otsu, Japan). The PCR parameters used were as follows: denaturation at 94°C for 7 min, followed by 30 cycles of denaturation (94°C for 1 min), annealing (50°C for 1 min), and extension (72°C for 1.5 min) and a final extension at 72°C for 10 min. The PCR products were purified with a PCR purification kit (Axygen, Tewksbury, MA, USA) and cloned into a pMD18-T vector (TaKaRa) to construct a methanogen 16S rRNA gene library. The clones were sequenced by BioSune Inc. (Beijing, China).

The 16S rRNA gene sequences were checked for chimeras with DECIPHER (11). Clones with ≥97% similarity were assigned as an operational taxonomic unit (OTU) using MOTHUR (12) based on the distance matrix. The methanogenic 16S rRNA gene sequences were then submitted to the GenBank database to search for homologous sequences using BLAST (13). The most similar sequences were retrieved and aligned using the ARB_EDIT4 tool in the ARB software package (14). A phylogenetic tree was constructed using neighbor-joining analysis (15), and the topology of the clustering was estimated with bootstrap sampling.

Methanogen strains and cultivation.

M. mazei Gö1^T was purchased from the Japan Collection of Microorganisms (JCM) (Tsukuba, Japan). Strain zm-15 was isolated from the Zoige wetland soil in this study and deposited in the China General Microbiological Culture Collection Center (CGMCC) (Beijing, China) under accession number CGMCC 1.5193.

For enrichment, soil samples were inoculated into basal medium supplemented with 20 mM (final concentration) methanol or acetate as the methanogenic substrate in an anaerobic chamber (Forma Anaerobic System 1029; Thermo Fisher Scientific, Waltham, MA, USA), as previously described (16). Complete media were dispensed into screw-cap serum bottles sealed with butyl rubber stoppers, with N₂ as the gas phase at 101.3 kPa. The enriched samples were incubated at 15°C for about 2 weeks prior to colony isolation via the Hungate rolling-tube procedure (17). The roll tubes were incubated at 15°C until single colonies appeared. Single colonies were picked, and the purified strain that produced CH₄ from methanol and acetate was designated strain zm-15. For identification of strain zm-15, the 16S rRNA gene was amplified with the universal archaeal primer A2F and the prokaryotic primer U1510R (see Table S1 in the supplemental material), as previously described (18). Both strains Gö1 and zm-15 were grown under anaerobic conditions in 50 ml of DSMZ 120 medium, as previously described (4), in a 100-ml serum bottle containing methanol or acetate (20 mM final concentration). Strain zm-15 formed large multicellular aggregates when grown in the medium, and the growth of cultures was determined from measurements of CH₄ production, as previously described (19). Cells at the exponential phase growing in acetate or methanol medium were collected at 5,000 × g for 15 min in an anaerobic chamber. After washing with prereduced phosphate buffer (1.7 mM cysteine-HCl · H₂O, 1.2 mM Na₂S · 9H₂O, 50 mM NaK-phosphate, pH 7.0; O₂ was removed from the buffer by 8 cycles of evacuation and flushing with N₂), the resting cells were prepared.

Methane determination.

Methane production was measured with a Shimadzu GC 14B gas chromatograph (Shimadzu, Kyoto, Japan) with a flame ionization detector and a C₁₈ column as described previously (20). The temperature parameters were set as follows: 50°C for the column, 80°C for the injector, and 130°C for the detector. N₂ was used as a carrier gas.

Enzymatic assays.

For the methanol-coenzyme M methyltransferase assay, strain zm-15 was cultured in 50 ml of DSM 120 medium with methanol as the sole carbon source until mid-exponential phase (the CH₄ concentration is about 4 mM, with methanol as the substrate), and then cells were harvested anaerobically at 5,000 × g for 15 min. The cell pellets were resuspended with 20 ml of wash solution (38 mM NaCl, 20 mM NaHCO₃, 9 mM NH₄Cl, 2 mM MgCl₂ · 6H₂O, 1.7 mM CaCl₂ · 2H₂O, 50 mM MOPS [morpholinepropanesulfonic acid], pH 7.0), collected by centrifugation at 7,400 × g for 15 min, and resuspended in 50 mM MOPS (pH 7.0). All the wash steps were performed anaerobically at room temperature. Buffers were prereduced before use. Cell extracts (CE) were prepared by sonication on ice (100 W; 1-s sonication and 2-s pause; 20 cycles) in an anaerobic chamber. The protein concentration was determined using Coomassie Protein Assay Reagent (Thermo Fisher Scientific). Assay mixtures were prepared anaerobically in 10-ml sealed vials, and the activity was determined as previously described (21).

For acetate kinase and phosphotransacetylase activity assays, 50 ml of mid-exponential-phase (the CH₄ concentration was about 5 mM, with acetate as the substrate) acetate-grown cultures of strain zm-15 in DSM 120 medium were centrifuged at 5,000 × g for 15 min. The cell pellets were washed aerobically with wash solution, centrifuged, and resuspended in lysis buffer (2 mM dithiothreitol [DTT], 100 mM Tris-HCl, pH 7.2). Then, the cells were lysed by sonication, and the protein concentration was determined. Acetate kinase activity was determined by the hydroxamate assay (22). Phosphotransacetylase activity was assayed by monitoring thioester bond formation, as previously described (23).

RNA extraction.

Strain zm-15 was grown in DSM 120 medium with 20 mM methanol or acetate until mid-exponential phase, and then cells were harvested. Total RNA was extracted by phenol-chloroform extraction, followed by isopropyl alcohol precipitation, as previously described (24). Total RNA was quantified by the NanoDrop Spectrophotometer (Thermo Fisher Scientific). Finally, 2 μg of each RNA sample was digested with 2 units of DNase I (Promega, Madison, WI, USA) at 37°C for 5 h to complete removal of genomic DNA.

RT-qPCR assay.

Reverse transcription (RT) reactions were performed using Moloney murine leukemia virus (MMLV) reverse transcriptase (Promega) according to the manufacturer's protocol with random primers (Promega) and 2 μg of DNase-treated total RNA as the template. The RT-generated cDNA was then used as the template, together with 25 μl SYBR green Premix (TaKaRa) and primers, as listed in Table S1 in the supplemental material. Real-time quantitative PCRs (qPCRs) were conducted with the Eppendorf Mastercycler system (Eppendorf, Hamburg, Germany), using a PCR program of one cycle of 95°C for 30 s, followed by 40 cycles of 95°C for 5 s, 52°C for 30 s, and 72°C for 30 s. A single sharp peak was produced for each PCR product with melting curve analysis, and transcript quantification was determined by the comparative threshold cycle (C_T) values. To estimate the copy numbers of the transcripts, the standard curve of each tested gene was generated by cloning the corresponding PCR fragment (100 to 200 bp) into the pMD-18T vector. The plasmid carrying the PCR fragment was then linearized at a site downstream of the target sequence, serially diluted, and used to generate the standard curve for quantitative PCR. The 16S rRNA gene, which was taken as a constitutively expressed housekeeping gene, was used as the biomass reference. The copy number of each gene was normalized against the 16S rRNA copies.

Determination of RNA transcript sequences at the 5′ and 3′ termini.

Total RNA was extracted from exponential-phase cultures of strain zm-15 and treated with DNase I. The 5′ and 3′ RNA termini were determined by the circularized-RNA RT-PCR (CRRT-PCR) protocol, as previously described (25). After denaturation at 70°C for 15 min, 10 μg of total RNA was self-ligated for circularization with T4 RNA ligase (Promega), T4 ligase buffer, and RNase inhibitor (Promega) in 25 μl at 37°C for 1 h. Then, the enzymes were removed by phenol chloroform extraction. RT-PCR was carried out with 0.5 pmol of the specific primers listed in Table S1 in the supplemental material, using MMLV reverse transcriptase and the circularized RNA as the template according to the manufacturer's instructions. The cDNA comprising the 5′-3′-ligated RNA was subsequently amplified with the gene-specific primer pair P1-P2, followed by a second PCR with the nested primers N1-N2 (see Table S1 in the supplemental material) and 0.4 to 0.6 kb amplification products of the first PCR as the template. KOD DNA polymerase (Toyobo, Osaka, Japan) was used for the amplification. The nested-PCR products of the 5′-3′-ligated RNA were cloned into a pMD-18T vector, and 24, 25, and 31 cDNA clones were sequenced for mtaA1, mtaC1B1, and the pta-ackA operon, respectively.

In vivo mRNA half-life assay.

Strain zm-15 was grown with methanol or acetate at 30°C or 15°C until mid-exponential phase, and then 100 μg/ml (final concentration) actinomycin D (MP Biomedicals) was added to inhibit transcription. Cells were collected after 0, 10, 20, 40, and 60 min, and total RNA was extracted and used for RT-qPCR. The primers used are listed in Table S1 in the supplemental material. The targets of the qPCR primer pairs are as follows: mtaA1F/mtaA1R, 3 to 121 nucleotides (nt) of the mtaA1 coding region; mtaC1F/mtaC1R, 519 to 653 nt of the mtaC1B1 coding region; ptaF/ptaR, 343 to 472 nt of the pta-ackA coding region. Quantification of the transcripts at different time points was normalized against the 16S rRNA copies and plotted on logarithmic scales. The half-life was calculated based on linear least-squares regression analysis, which required a 50% decrease in the initial transcript abundance.

In vitro half-life assay for mRNA mutants.

All mRNA transcripts were generated by in vitro transcription for the tested genes from a linearized plasmid. To construct the linearized plasmid, the PCR product of a given mutant transcript was cloned into vector pSPT19. For the hybrid transcription template, overlapping PCR was performed as previously described (26). KOD DNA polymerase was used in the amplification reaction with the corresponding specific primers listed in Table S1 in the supplemental material. The in vitro transcription was performed using an RNA synthesis kit with T7 RNA polymerase (Roche, Basel, Switzerland) according to the manufacturer's instructions. The in vitro transcripts were treated with DNase I and purified by isopropyl alcohol precipitation.

CE from mid-exponential growth phase cultures of strain zm-15 were used as the crude nucleases for the mRNA stability assay (27). Cultures were harvested at 5,000 × g for 15 min to pellet cells, and the cells were washed with washing solution (38 mM NaCl, 20 mM NaHCO₃, 9 mM NH₄Cl, 2 mM MgCl₂ · 6H₂O, 1.7 mM CaCl₂ · 2H₂O, 50 mM MOPS, pH 7.0). The cells were then repelleted and resuspended in HEPES buffer (100 mM NaCl, 1 mM DTT, 20 mM HEPES, pH 7.5) with glycerol (10% [vol/vol]) and lysed by sonication. The protein concentration was determined with Coomassie Protein Assay Reagent.

Before the reaction, purified in vitro transcripts were denatured at 90°C for 1 min and renatured for 15 min at 30°C or 15°C to obtain mRNA structure identical to the that of natural transcript at moderate or low temperatures (28). CE was treated with DNase I at 37°C for 15 min to remove the cellular DNA. The in vitro mRNA stability assays were carried out in 10 μl HEPES buffer containing the synthetic mRNA (500 ng) and crude nucleases (1 μg protein) at 30°C. The mRNA decay reaction was terminated at −80°C by freezing the mixture immediately in an ultralow-temperature freezer (Thermo Fisher Scientific). Next, the reaction mixture was run on a 1% agarose gel and stained with ethidium bromide. The remaining mRNA was determined by analyzing the scanned-RNA band density with TotalLab Quant software (TotalLab, Newcastle, United Kingdom), and the in vitro half-life was calculated from the linear least-squares regression of the logarithm of the RNA band density against the time of CE incubation.

Nucleotide sequence accession numbers.

The methanogenic 16S rRNA gene sequences for diversity analysis and strain zm-15 were submitted to the GenBank database under accession numbers KF360007 to KF360023. The genes involved in methanol-derived and aceticlastic methanogenesis in M. mazei zm-15 acquired in this study were sequenced. The sequences were identical to those of the genes in M. mazei Gö1, i.e., mtaA1 (MM1070), mtaA2 (MM0176), mtaB1 (MM1647), mtaB2 (MM1074), mtaB3 (MM0175), mtaC1 (MM1648), mtaC2 (MM1073), mtaC3 (MM0174), pta (MM0496), and ackA (MM0495).

RESULTS

Isolation of psychrotolerant M. mazei zm-15 prevalent in the cold Zoige wetland.

To explain the mechanisms of cold adaptation of methanol-derived methanogenesis, which is prevalent in the cold Zoige wetland, a wetland-predominant methanogen that performed both methylotrophic and aceticlastic methanogenesis was isolated. The isolate, M. mazei strain zm-15, shared 100% 16S rRNA gene similarity with the predominant clone, ZW-M-4, in the methanogen 16S rRNA library of the wetland soil (see Fig. S1 in the supplemental material) and 99.6% similarity with that of M. mazei Gö1. Moreover, unlike M. mazei Gö1, which grows at 30 to 40°C and cannot grow at 15°C, strain zm-15 grows at 8 to 37°C and optimally at 30°C. Therefore, it appears to be a psychrotolerant strain of M. mazei.

Methanol-derived methanogenesis is more cold adaptive than aceticlastic methanogenesis in M. mazei zm-15.

To compare the cold sensitivity of methylotrophic and aceticlastic methanogenesis, strain zm-15 was grown with methanol or acetate at 30°C or 15°C. Methane production was measured during the entire growth phase. As shown in Fig. 1, at either growth temperature, methane production rates were higher in the methanol cultures (0.0173 ± 0.0005 h⁻¹ at 30°C and 0.0057 ± 0.0007 h⁻¹ at 15°C) than in the acetate cultures (0.0108 ± 0.0001 h⁻¹ at 30°C and 0.0014 ± 0.0001 h⁻¹ at 15°C). This can be partially attributed to the favorable thermodynamics of methanol-derived methanogenesis (5). Remarkably, the rate of aceticlastic methanogenesis was much more temperature sensitive than that of methylotrophic methanogenesis and declined 7.7-fold between 30°C and 15°C. In contrast, the rate of methanol-derived methanogenesis decreased only 3-fold. Temperature-related differences between methylotrophic and aceticlastic methanogenesis rates were even greater in resting cells (see Fig. S2 in the supplemental material). The former remained almost unchanged at 15°C versus 30°C, while the rate of aceticlastic methanogenesis was barely detectable at 15°C. Moreover, strain zm-15 produced methane from methanol at 8 to 10°C, while aceticlastic methanogenesis occurred only above 15°C, and no methane production from acetate was observed at 10°C over more than 6 months. These findings suggest that methanol-derived methanogenesis is more cold adaptive than aceticlastic methanogenesis in zm-15.

FIG 1.

CH₄ production during the growth of M. mazei zm-15 with methanol (A) or acetate (B) at 30°C (▲) versus 15°C (■). The data are means from three replicates of independent cultures ± standard deviations. The arrows indicate the mid-exponential phase of zm-15.

Expression of the mta genes was less cold sensitive than that of the genes for aceticlastic methanogenesis.

To discover whether the two pathways respond to low temperature mostly at the mRNA level, the genes specific to methanol- and acetate-derived methanogenesis were first determined. Based on the fact that M. mazei Gö1 carries mtaA1 and mtaA2, and mtaC1B1, mtaC2B2, and mtaC3B3 for three isomers of methanol methyltransferase, by using the specific DNA fragments as primers, the orthologs were all amplified from the zm-15 genome through PCR. Using RT-qPCR, the mRNA abundances of eight methanol-derived methanogenesis-related genes and the ackA, pta, and cdh genes involved in acetate-derived methanogenesis were detected in each substrate-grown culture. As shown in Table S2 in the supplemental material, ackA and pta, which encode enzymes acting in acetate activation, were greatly induced by acetate. While mtaA1 and mtaC1B1 were significantly induced by methanol, mtaA2 and mtaC3B3 were severely depressed by methanol, whereas mtaC2B2 exhibited similar mRNA levels in methanol and acetate, similar to a finding in M. mazei Gö1 (4). This suggests that the enzyme complex encoded by mtaA1 and mtaC1B1 plays the main role in methanol-derived methane production. Subsequently, temperature-related mRNA abundance assays for the genes involved in the two pathways were performed on the corresponding substrate-grown cultures, and only mtaA1 and mtaC1B1 were chosen for the methanol-derived methanogenesis pathway. Table 1 shows that the mRNA abundances of the three genes encoding the methanol-CoM methyltransferase complex (Mta) were <2 times higher in the 30°C culture than in the 15°C culture, while the mRNA levels of ackA and pta were 4.5 and 6.8 times higher in the 30°C than in the 15°C culture. The activities of the enzymes involved in aceticlastic methanogenesis were also reduced more than those for methanol-derived methanogenesis in 15°C-grown cultures (see Table S3 in the supplemental material). This indicated that the cold adaptation of the two pathways may be at the mRNA level, namely, mtaA1 and mtaC1B1 expression was more cold adaptive than that of ackA and pta at the transcriptional level. A recent proteomics study (29) also showed the upregulation of the MtaC protein in the 15°C culture of Methanosarcina barkeri.

TABLE 1.

Levels of mRNAs key to methanol-derived and aceticlastic methanogenesis in M. mazei zm-15 at moderate and low temperatures

Gene	Copy numbera		Fold change (30°C/15°C)
	30°C	15°C
mtaA1	64.53 ± 1.53	38.69 ± 1.57	1.67
mtaB1	128.02 ± 3.45	81.14 ± 1.89	1.58
mtaC1	156.29 ± 4.35	82.73 ± 3.10	1.89
ackA	69.21 ± 4.92	15.38 ± 1.66	4.50
pta	121.97 ± 3.41	18.04 ± 2.09	6.76

^aThe numbers were calculated from the gene copy numbers/10⁵ 16S rRNA copies. The values are the means ± standard deviations from three replicates.

mtaA1 and mtaC1B1 transcripts possessed high stabilities at both temperatures, while the pta-ackA transcript possessed reduced stability at low temperatures.

To elucidate whether the different cold-responsive mRNA abundances of mtaA1 and mtaC1B1 compared with ackA and pta were attributed to cold-induced transcription or mRNA degradation, the genes' organization and their promoters in zm-15 were determined through RT-PCR (see Fig. S3 in the supplemental material). As shown in Fig. 2, mtaA1, mtaC1 plus mtaB1, and pta plus ackA constituted three separate operons. Next, using RT-qPCR, the in vivo half-lives of mtaA1, mtaC1B1, and pta-ackA transcripts were determined in the 30°C and 15°C cultures after inhibiting transcription with 100 μg/ml actinomycin D according to the method of Hennigan and Reeve (30). The results showed that mtaA1 and mtaC1B1 were very stable in the cultures grown at both temperatures, with half-lives of about 1 h. In contrast, the half-life of pta-ackA was relatively short (25 min) at 30°C and even shorter (15.5 min) at 15°C (Fig. 3 and Table 2). This indicated that transcript stability contributed, at least partially, to the cold-responsive differential mRNA levels between the key genes for methanol- and acetate-derived methanogenesis.

FIG 2.

Organization of genes for methanol-CoM methyltransferase (mtaA1 and mtaC1B1), acetate kinase (ackA), and phosphotransacetylase (pta) in M. mazei zm-15. The gray arrows show the genes' coding regions and orientations, the bent arrows indicate the TSS, and the numbers indicate the nucleotides between the TSS and initial codon. The arrowheads point to the stop sites for transcription, and the mtaA1, mtaC1B1, and pta-ackA transcripts possess 90-nt, 29-nt, and 43-nt 3′ UTRs, respectively. The thin arrows refer to the intergenic spacers with RT-PCR products.

FIG 3.

Stabilities of mRNAs for methylotrophic and aceticlastic methanogenesis genes. The percentages of the mRNAs of mtaA1 (A), mtaC1B1 (B), and pta-ackA (C) operons remaining in strain zm-15 cultured at 30°C (▲) and 15°C (■) were determined by RT-qPCR. At time zero, 100 μg/ml actinomycin D was added to the cultures. The data are means from three replicates of independent cultures ± standard deviations.

TABLE 2.

In vivo half-lives of mRNAs for mta and pta-ackA in 30°C- and 15°C-cultured M. mazei zm-15

Transcript	Half-life (min)a		Fold change (30°C/15°C)
	30°C	15°C
mtaA1	61.66 ± 7.03	59.75 ± 5.11	1.03
mtaC1B1	56.45 ± 4.50	58.38 ± 2.78	0.97
pta-ackA	25.13 ± 0.58	15.48 ± 2.48	1.62

^aHalf-lives were calculated by linear least-square regression analysis of the transcript abundances at different time points. The values are means ± standard deviations from three replicates.

mtaA1 and mtaC1B1 mRNAs have large 5′ UTRs.

Most M. mazei Gö1 transcripts possess long 5′ untranslated regions (UTRs) (31), including the three operons of mtaCB of Methanosarcina acetivorans C2A (32). To determine whether the mRNA stability is attributable to the transcript architecture, the transcription start sites (TSS) and sequences of the 5′ UTRs and 3′ UTRs of mtaA1, mtaC1B1, and pta-ackA were determined by CRRT-PCR. Similar to the M. mazei Gö1 and M. acetivorans C2A transcripts, large 5′ UTRs of 270 and 238 nt were detected in the mtaA1 and mtaC1B1 mRNAs of zm-15, while only a short 27-nt 5′ UTR was found in the pta-ackA transcript (Fig. 2). Through sequence alignment (see Fig. S4 in the supplemental material), we found that the mtaA1 5′ UTR of zm-15 shared 100% sequence identity with that of M. mazei Gö1 and 83.3% similarity with that of M. acetivorans C2A. The mtaC1B1 5′ UTR of zm-15 showed 97.9% similarity to that of M. mazei Gö1 and 71.9% similarity to that of M. acetivorans C2A. Upstream of the predicted ribosome binding site (RBS), the two 5′ UTRs are A/T rich, especially the mtaA1 5′ UTR. Also, 90-nt, 29-nt, and 43-nt 3′ UTRs were found in mtaA1, mtaC1B1, and pta-ackA transcripts, respectively (Fig. 2), all of which were U rich (data not shown). Hence, transcripts with large 5′ UTRs may be common in methanogenic archaea.

The large 5′ UTRs significantly contribute to mtaA1 and mtaC1B1 mRNA stability.

To test the contributions of the 5′ UTRs of mtaA1 and mtaC1B1 to their mRNA stability, leaderless transcripts of the two genes were constructed by in vitro transcription. The in vitro half-lives were determined by measuring the remaining mRNAs after digestion with CE of 30°C-cultured zm-15 cells for up to 1 h. The results indicated that removal of their intrinsic 5′ UTRs reduced the half-lives of mtaA1 and mtaC1B1 transcripts by 25% and 32%, respectively (Fig. 4). Furthermore, the mutant transcripts were even less stable at 15°C (53% and 42%, respectively), indicating the special contribution of the 5′ UTR to maintaining mRNA stability.

FIG 4.

Effect of temperature on the stabilities of mtaA1 and mtaC1B1 transcripts in vitro. The transcripts were renatured at 30°C (A and B) or 15°C (C and D) and then incubated with zm-15 CE at 30°C for different times. (A and C) The remaining mRNAs of leaderless and wild-type mtaA1 and mtaC1B1 treated with CE were visualized on agarose gels. −, CE without mRNA; +, mRNA without CE; black arrows, coding region; gray rectangles, 5′ UTR. (B and D) Regression curves of mRNA degradation. ▲, leaderless mtaA1; ■, wild-type mtaA1; ⧫, leaderless mtaC1B1; ●, wild-type mtaC1B1.

Furthermore, hybrid pta transcripts were constructed by fusion of the 5′ UTR from mtaA1 or mtaC1B1 to the leaderless pta mRNA through in vitro transcription, and the half-lives were measured using a procedure similar to that used for mta transcripts. As shown in Fig. 5, addition of the mtaA1 and mtaC1B1 5′ UTRs prolonged the half-lives of the pta-ackA transcript mutants that were renatured at 30°C by 2.5- and 1.8-fold, respectively. The half-lives were prolonged even more (3.2- and 2.5-fold, respectively) when the transcripts were renatured at 15°C. This confirms the role of the 5′ UTR in transcript stability, especially in cold stability.

FIG 5.

Effect of temperature on stability of pta-ackA transcripts in vitro. The transcripts were renatured at 30°C (A and B) or 15°C (C and D) and then incubated with zm-15 CE at 30°C for different times. (A and C) The remaining mRNAs of leaderless and wild-type pta-ackA and pta-ackA fused with the 5′ UTR of mtaA1 or mtaC1B1 treated with CE were visualized on agarose gels. −, CE without mRNA; +, mRNA without CE; black arrows, coding region; gray rectangles, 5′ UTR. (B and D) Regression curves of mRNA degradation. ▲, leaderless pta-ackA; ■, pta-ackA fused with wild-type 5′ UTR; ◆, pta-ackA fused with mtaA1 5′ UTR; ●, pta-ackA fused with mtaC1B1 5′ UTR.

DISCUSSION

Temperature is one of the important determinants of methanogenic pathways and methanogen populations in ecosystems. The contributions of aceticlastic methanogenesis in lower-temperature environments have been reported in rice field soil (33), lake sediment (34), and permafrost soil (35). However, we found a methanol-derived methanogenesis rate higher than that from acetate in the cold Zoige wetland soil, and methanol supported an even higher methanogenesis rate at 15°C than at 30°C (3). The molecular basis of the cold activity of methanol-derived methanogenic pathways was investigated in M. mazei zm-15. We conclude that the transcript cold stability of the essential genes contributes to the higher activity of the methylotrophic pathway and that the large 5′ UTR plays a significant role in the cold stability of these transcripts.

It has been determined that the mRNA stability in Saccharomyces cerevisiae is affected by the poly(A) tail length at the 3′ UTR and the m7G cap at the 5′ UTR (36). In higher organisms, mRNA stability is mainly regulated by the elements embedded in the transcript 3′ UTR (37, 38). In contrast, in bacteria, the 5′-terminal stem-loop structures can protect transcripts from degradation by RNase E (39), resulting in more stable mRNA. E. coli ompA mRNA is stabilized by its long, 133-nt 5′ UTR (7, 40). In the present study, large 5′ UTRs contributed to the mRNA stability of methanol-derived methanogenesis genes in M. mazei zm-15. The impact of a large 5′ UTR on mRNA stability can be attributed to the mode of mRNA degradation. The sensitivity to endonuclease E in Escherichia coli, a protein essential for mRNA decay and processing, depends on the 5′ termini of RNAs (41, 42). In addition, higher-order structures of the 5′ UTR affect translation by facilitating ribosome binding to the mRNA, which also masks the RNase E cleavage site, thus protecting the mRNA from degradation (43). Though the mechanism of mRNA decay is not yet known for methanogenic archaea, RNA processing is through endonucleolysis in Methanocaldococcus jannaschii, as determined by 3′ rapid amplification of cDNA ends (RACE) and 5′ RACE analysis (44). However, no characteristic sequence surrounding the cleavage sites has been found, except for an AUG translation start codon and, in most cases, a ribosome binding site.

The 5′ UTR of a transcript is predicted to more specifically sense the ambient temperature based on temperature-sensitive base pair formation (45). Using the Mfold Web Server (46), different potential secondary structures of mtaA1 and mtaC1B1 5′ UTRs were predicted (see Fig. S5 in the supplemental material). The large 5′ UTR (159 nt) of the cold shock protein A (CspA) mRNA in E. coli undergoes a temperature-dependent higher-structure rearrangement, thus functioning as an RNA thermometer. cspA mRNA exhibits a cold shock stability shift and modulates CspA translation (47). Additional functions of the methanogenic transcript 5′ UTRs have been reported. The large 5′ UTR of cdh, encoding ACS/CODH, functions in transcription pretermination of the gene in Methanosarcina thermophila (48). Moreover, the large 5′ UTRs are predicted to play multiple roles. They are the target elements of noncoding regulation RNAs by cis- or trans-actions (49) and are the essential elements of riboswitches (50).

In conclusion, this study demonstrated that in the cold-adaptive M. mazei zm-15, the transcripts of methanol-CoM methyltransferease are more stable at cold temperatures, and the 5′ UTR determined the cold stability. The cold stability of the mRNAs may confer cold activity of methanol-derived methane production, but not aceticlasitc methanogensis performed in a single strain. This work also provided an example of the significance of transcript stability in gene regulation. In contrast to halophilic Euryarchaeota and Crenarchaeota, in which the leaderless transcripts are dominant, posttranscriptional regulation can play significant roles in methanogenic archaea with the prevalence of large 5′ UTR transcripts.