|Home | About | Journals | Submit | Contact Us | Français|
In vivo expression of CO dehydrogenase/acetyl coenzyme A synthase in Methanosarcina spp. is coordinately regulated in response to substrate by at least two mechanisms: differential transcription initiation and early elongation termination near the 3′ end of a 371-bp leader sequence. This is the first report of regulation of transcription elongation in the Archaea.
Methanogenic species in the genus Methanosarcina are the most metabolically diverse among methanogens, with the ability to grow by CO2 reduction with H2, methyl reduction with H2, aceticlastic fermentation of acetate, or methylotrophic catabolism of methanol, methylated amines, and dimethyl sulfide (34). Methanosarcina acetivorans is also reported to grow with carbon monoxide (23). These organisms preferentially use substrates with higher free energy, such as methanol and trimethylamine (TMA), before utilizing acetate, exhibiting diauxic growth in the presence of both types of substrates (4, 10, 16, 21, 26). Despite their preference for nonaceticlastic substrates and the relatively lower growth rates and yields of Methanosarcina spp. with acetate, approximately 70% of the biogenic methane produced globally is generated from the catabolism of acetate (19). As a result, regulation of genes involved in the aceticlastic pathway is likely to have a significant impact on anaerobic degradation processes and the global biogenesis of methane.
Several enzymes associated with aceticlastic methanogenesis, including acetate kinase, phosphotransacetylase, carbonic anhydrase, and carbon monoxide dehydrogenase/acetyl coenzyme A (acetyl-CoA) synthase (CODH/ACS), are regulated in response to substrates (13, 14, 17). Expression of the CODH/ACS operon (cdhABCDE), which catalyzes the catabolism of acetyl-CoA in the aceticlastic pathway, is highly regulated in response to growth on acetate compared with methanol and TMA, but the mechanism(s) of this regulation is not known (28, 31). Two CODH/ACS operons are detected in the annotated genome sequences of M. acetivorans (11), Methanosarcina barkeri (20), and Methanosarcina mazei (7). In contrast, only one cdhABCDE operon has been detected in Methanosarcina thermophila (12). In M. acetivorans, both of these operons appear to be regulated in response to substrate, although the extent of the regulation is greater in one (MA3680) than in the other (MA1016) (17). Northern analysis of CODH/ACS mRNA from Methanosarcina thermophila indicates that the regulation of this CODH/ACS ortholog in response to substrate occurs, at least in part, at the level of mRNA (28). The regulation of this ortholog was corroborated with translational fusions of the M. thermophila CODH/ACS promoter to lacZ (2) and by peptide fragment analysis of a CODH/ACS component, CdhA, in M. mazei (9). This CODH/ACS operon has an unusually long 371-bp 5′ leader region (28). In contrast, the 5′ leader sequence of CODH/ACS in the obligately aceticlastic Methanosaeta concilii (formerly “Methanothrix soehgenii”) is less than 100 bases long and has no significant sequence similarity to the methanosarcinal 5′ leader regions (8), such as the one found in M. thermophila. These observations suggest that the CODH/ACS 5′ leader region has a regulatory role, possibly at the transcriptional and/or posttranscriptional level.
The purpose of this study was to localize regulatory regions that mediate expression of the catabolic CODH/ACS complex in the aceticlastic Methanosarcina to determine whether the 5′ leader region is involved in expression. The region upstream of the transcriptional start site and leader sequence from the M. thermophila TM1 CODH/ACS operon was fused to lacZ as a translational reporter to study differential gene expression on different substrates. Deletion analyses of sequences adjacent to the promoter combined with quantitative reverse transcriptase PCR (qRT-PCR) analyses of the transcript indicate that CODH/ACS expression is regulated in part by the sequence located downstream of the CODH/ACS promoter in the 5′ leader in response to acetate and methylotrophic substrates.
The current study focused on the CODH/ACS ortholog in M. thermophila (GenBank accession No. L20952), an ortholog of M. acetivorans MA3860, M. barkeri MbarA0204, and M. mazei MM0684, which had been confirmed previously to be highly regulated in response to substrate (2, 9, 28). Alignment of the cdhABCDE mRNA leader sequence from the transcriptional start site to the beginning of the genes encoding CODH/ACS revealed that the orthologous 5′ leader region sequences between Methanosarcina spp. were highly conserved, ranging from 73% to 87% sequence identity, which is similar to the sequence conservation of 77% to 86% identity observed between methanosarcinal cdhA encoding orthologs (Fig. (Fig.1).1). In contrast, only 45% to 62% sequence identity is observed within the region 256 bp upstream of the transcription start site. These regions were evaluated further by deletion analysis to determine whether the highly conserved leader sequence has a role in CODH/ACS regulation.
In a prior report that employed a translational fusion of the cdh promoter to lacZ, CODH/ACS from M. thermophila TM1 (DSM 1825) was shown to be upregulated up to 54-fold or 31-fold during growth on acetate relative to growth on methanol or TMA, respectively (2). To identify sequences involved in CODH/ACS regulation, a series of deletions were generated in the CODH/ACS lacZ reporter construct upstream of the transcriptional start site and downstream in the 371-bp 5′ leader region between the transcription start site and the translation start site of the CODH/ACS operon (Fig. (Fig.2).2). Progressive deletions were made from position 751 to within 262 and 71 bases upstream of the transcription start site to detect both distant and near putative cis-regulatory elements. To detect putative cis-regulatory elements in the 5′ untranslated region (UTR), progressive downstream deletions were made immediately upstream of the predicted ribosomal binding site of the cdhABCDE operon to within 211, 133, and 14 bases of the transcription start site. Detailed deletion strategies are provided in Tables S1 and S2 in the supplemental material. All deletions retained the wild-type ribosomal binding site of cdhA. M. acetivorans transformed with either the wild-type reporter plasmid pEA103 or deletion plasmids were inoculated (10% [vol/vol]) in triplicate into liquid medium containing 0.1 M sodium acetate, TMA, or methanol (27) and incubated to mid-exponential growth phase at 35°C. Cultures were sampled to perform β-galactosidase assays as described previously (2). Previous experiments (2) demonstrated that the copy number of pEA103 in M. acetivorans is the same when grown on the three different substrates.
The effects of deletions upstream of cdhA on the expression of β-galactosidase in response to different substrates are shown in Table Table1.1. Deletion of DNA sequences upstream of the promoter (Δ62 or Δ63 construct) did not appear to have an effect on expression of β-galactosidase compared with the wild-type sequence in cells grown with all substrates, indicating that upstream cis elements were not involved in regulation of transcription initiation in response to substrate. The Δ64 construct resulted in extremely low β-galactosidase levels, which is expected since the deleted sequence included the CODH/ACS promoter. Expression with the Δ65 construct, which had 124 bases deleted near the 3′ end of the leader sequence, was not significantly different from the wild type when cells were grown on TMA. There was a difference, however, when cells were grown on methanol and on acetate. Some deletions downstream of the transcription start site within the 5′ leader sequence did have an effect on expression during methylotrophic growth. Table Table11 shows that Δ66 and Δ67 constructs had only four- and 13-fold differences in expression, respectively, between acetate- and methanol-grown cells, compared to the 61-fold difference observed in wild-type constructs. The same was true for cells grown on TMA, where the Δ66 and Δ67 constructs had differences in expression that were seven- and fivefold higher, respectively, compared to a 20-fold difference for the wild-type construct. However, the relative expression between acetate and each of the two methylotrophic substrates was different between the Δ66 and Δ67 constructs, which suggests a difference in regulation by these substrates. Another study also noted similar differences in response between different substrates to deletions in the 5′ leader sequence of methylamine methyltransferase mtaCB operons in M. acetivorans (5). The authors suggest this could be due to two different trans factors within this region. Further investigation of the +28 and +133 region is necessary to determine if different trans elements or cis elements, such as the conserved poly(U) sequence, have a role in the observed differential regulation between TMA- and methanol-grown cells. It should also be noted that there is a possible open reading frame that begins at position +296 on the opposite strand as cdhA and that terminates at the site of transcription initiation. Further investigation of this putative open reading frame is necessary to determine if it plays a role in the differential regulation observed. Regardless of the differential deletion effects observed between methylotrophic substrates, these data clearly indicate that the 5′ leader region is involved in the regulation of expression based on the substrate, and that the sequence within +28 to +211 bp of the 5′ leader sequence has a critical role in the regulation of CODH/ACS expression.
In bacterial systems, long 5′ UTRs are involved in transcriptional regulation via multiple mechanisms, usually via formation of secondary structures within the RNA. One example of this type of regulation is the attenuation mechanism, which involves the formation of stem-loop structures, along with a long stretch of uridines immediately following one of the structures (25, 35). The results presented in this paper do not rule out the possibility that conformational changes in mRNA secondary structure of the 5′ leader region have a role in regulation; however, further investigation is needed to determine its precise role in regulation. Using secondary structure analysis, the leader region of cdhA has multiple putative secondary structures predicted, with a free energy ΔG of −39 kcal/mol. In the majority of the predicted structures, the 3′ end of the leader appears to form a relatively strong hairpin structure. Conformational changes can occur due to the activity of regulatory proteins or other compounds, such as ligands, binding to the RNA. More experiments are necessary to determine what other RNA conformations could be occurring and what factors could be causing the changes in structure. Conformational change in mRNA secondary structure causes early transcription termination, as well as affecting translation. Previous studies of cdh have shown this operon to be regulated at the level of transcription, but it was not known whether regulation was also occurring at the level of translation.
RNA was extracted from wild-type M. acetivorans harvested during mid-exponential growth on either acetate or methanol with the RNeasy kit (Qiagen) following the manufacturer's instructions. RNA (50 ng) was used as the template for each qRT-PCR, using the iScript one-step RT-PCR kit (Bio-Rad), following the manufacturer's instructions. The primers used in the qRT-PCRs are listed in Table S1 in the supplemental material. To compare the differences in mRNA levels, the threshold cycle (CT) values for each reaction were normalized to the CT value for the gene product of MA4504 (33). The differences were calculated using the formula 2−ΔΔCT, where ΔΔCT is the difference between the normalized CT values of acetate-grown cultures and those of the methanol-grown cultures (24). The difference in mRNA levels in wild-type cells grown with acetate and methanol determined by qRT-PCR was 68 ± 20-fold, compared with the 62 ± 6-fold difference in β-galactosidase level expressed by the M. acetivorans strain containing the pEA103 cdhA′::lacZ reporter plasmid. Using qRT-PCR with primers UTRFor and LacZRev, similar differences in β-galactosidase levels were also observed in Δ65 and Δ66 plasmid constructs compared with the wild type during growth with methanol and TMA (data not shown). The difference between transcript and protein levels in both the wild type and deletion mutants was not significant, indicating that translation is not a factor in the regulation of CODH/ACS expression.
In bacterial systems, some 5′ UTRs regulate by differential transcript stability attributed to altered secondary structures within the leader sequence (1, 32). To determine whether differential transcript stability was a viable mechanism for methanosarcinal CODH/ACS regulation, transcription was inhibited, and transcript decay rates were compared between cells grown on different substrates. Actinomycin D was shown previously to inhibit transcription in halophilic and thermophilic Archaea (3, 15). We confirmed that actinomycin D effectively inhibited transcription in M. acetivorans by showing that actinomycin D inhibited incorporation of the 3H-uridine exponentially growing cultures as a result of transcription inhibition, while cells without actinomycin D continued to show 3H-uridine incorporation (data not shown).
Samples were taken from exponential-phase cultures of M. acetivorans grown with either acetate or methanol at 0, 2, 5, 10, 15, and 30 min after addition of 100 μg/ml actinomycin D, and RNA was extracted as described above. The RNA was quantified and used as a template for qRT-PCR using primers UTRFor and 405Rev (see Table S2 in the supplemental material). These primers were specific for the entire 5′ leader sequence as well as 40 bp of the cdhA structural gene, generating a product that was 405 bp in length. No significant difference (P < 0.05) was observed between the ratio of CT values of mRNA in acetate-grown cells versus those of methanol-grown cells at any of the time points, indicating that the relative stabilities of the message isolated from acetate- and methanol-grown cells were similar (data not shown).
To determine whether differential elongation within the 5′ leader was a possible mechanism of regulation, RT-PCR products generated with nine primers that hybridized along the 5′ length of the RNA transcript were quantified by qRT-PCR to detect any changes in transcript length between acetate- and methanol-grown cells (Fig. (Fig.3).3). For this experiment, one forward primer (UTRFor) was used in combination with nine different reverse primers (see Table S1 in the supplemental materials), generating PCR products of various lengths (Fig. (Fig.3).3). The template used for the qRT-PCRs was RNA isolated from wild-type M. acetivorans harvested during mid-exponential growth on either acetate or methanol. The difference in the amounts of transcript expressed between acetate- and methanol-grown cells was calculated as described above for each combination of primers (Fig. (Fig.3).3). A 10- to 15-fold difference in transcript level was observed within 358 bases of the 5′ end in cells grown on acetate and methanol. However, transcript levels 405 bases downstream of the 5′ end showed a significantly greater difference, increasing to a 68-fold difference between acetate- and methanol-grown cells. These results suggest that termination of elongation occurred between 358 and 405 bases downstream of the 5′ end of the cdh transcript during methylotrophic growth. Another possible cause for the change in transcript level could be differential processing of mRNA. While the actinomycin D experiment ruled out differential mRNA decay, it could not rule out the possibility of differential processing of mRNA, as the addition of actinomycin D could have disrupted a processing event. The mRNA may be cleaved between bp 358 and 405 during methylotrophic growth, causing the difference in transcript levels between acetate and methylotrophic growth. In addition to the difference noted between bp 358 and 405, a 10-fold difference in transcript levels between cells grown on acetate and those grown on methanol was observed from +70 to +358 bases downstream of the transcription start site. This observation indicates that differential transcript initiation also affects CODH/ACS expression.
In bacterial systems, long 5′ UTRs are involved in transcriptional regulation by either inhibiting or promoting transcription elongation (36). Mechanisms such as attenuators, discussed above, and riboswitches involve the formation of different secondary structures within RNA. There are multiple permutations of these systems present in both eukaryotes and prokaryotes. Some systems involve a regulatory protein (25) or ligands (22, 29, 30) that bind to the RNA, which creates a conformational change that either terminates or allows transcription to occur. Termination of transcriptional elongation in sequences distal to the regulatory CODH/ACS 5′ leader sequence is consistent with regulatory mechanisms involving changes in secondary structure, such as attenuation or riboswitches. Another possible mechanism for termination of transcription would involve factor-dependent termination, such as that seen in bacterial systems involving the termination factor Rho. Under methylotrophic substrates, this termination factor might bind to the RNA, translocating along the RNA until it encounters the transcription complex and causes transcription to terminate. Under aceticlastic conditions, a conformational change in the RNA would prevent termination by the terminating protein. This model is also consistent with the result of this study.
5′ leader regions identified within the Archaea include the 113-bp leader region identified upstream of a DEAD-box RNA helicase in the Antarctic methanogen Methanococcoides burtonii (18) and the methyltransferase genes in Methanosarcina spp. (6). In both of these examples, the UTR was implicated in regulation, although the role of the leader region in regulation was not confirmed. The system described in this paper appears to involve sequences downstream of the start of transcription, well into the 5′ leader sequence. Our results also indicate that the 5′ leader region has a role in the regulation of CODH/ACS postinitiation via termination of elongation during methylotrophic growth. This conclusion is supported by several observations. First, posttranscriptional regulation by differential translation was ruled out, as the difference in the protein levels and the transcript levels were not significantly different. Second, cdh transcript stability levels were similar in cells grown under aceticlastic and methylotrophic conditions, suggesting that differential mRNA degradation is not likely. Finally, a significant difference in transcript levels was observed 405 bases downstream of the 5′ end of the transcript. The results indicate that methanosarcinal CODH/ACS expression is controlled by at least one mechanism at the level of transcription elongation as part of the regulatory strategy employed by these methanogenic Archaea to efficiently direct carbon and electron flow in anaerobic consortia during fermentative processes. To the best of our knowledge, this is the first evidence of regulation at the level of transcriptional elongation by a 5′ leader region as a mechanism for gene regulation in the Archaea. Future studies will focus on identification of trans-acting elements and putative secondary structures to characterize the paradigm for catabolic CODH/ACS regulation in these Archaea.
We thank Zvi Kelman, Rob Gunsalus, and Harold Schreier for critical review of the manuscript.
This work was supported in part by grants from the Department of Energy, Energy Biosciences Program (DE-FG02-93ER20106), and the National Science Foundation, Division of Cellular and Bioscience (MCB0110762).
Published ahead of print on 18 September 2009.
†Supplemental material for this article may be found at http://jb.asm.org/.