Characterization of an Archaeal Two-Component System That Regulates Methanogenesis in Methanosaeta harundinacea

Two-component signal transduction systems (TCSs) are a major mechanism used by bacteria in response to environmental changes. Although many sequenced archaeal genomes encode TCSs, they remain poorly understood. Previously, we reported that a methanogenic archaeon, Methanosaeta harundinacea, encodes FilI, which synthesizes carboxyl-acyl homoserine lactones, to regulate transitions of cellular morphology and carbon metabolic fluxes. Here, we report that filI, the cotranscribed filR2, and the adjacent filR1 constitute an archaeal TCS. FilI possesses a cytoplasmic kinase domain (histidine kinase A and histidine kinase-like ATPase) and its cognate response regulator. FilR1 carries a receiver (REC) domain coupled with an ArsR-related domain with potential DNA-binding ability, while FilR2 carries only a REC domain. In a phosphorelay assay, FilI was autophosphorylated and specifically transferred the phosphoryl group to FilR1 and FilR2, confirming that the three formed a cognate TCS. Through chromatin immunoprecipitation–quantitative polymerase chain reaction (ChIP-qPCR) using an anti-FilR1 antibody, FilR1 was shown to form in vivo associations with its own promoter and the promoter of the filI-filR2 operon, demonstrating a regulatory pattern common among TCSs. ChIP-qPCR also detected FilR1 associations with key genes involved in acetoclastic methanogenesis, acs4 and acs1. Electrophoretic mobility shift assays confirmed the in vitro tight binding of FilR1 to its own promoter and those of filI-filR2, acs4, and mtrABC. This also proves the DNA-binding ability of the ArsR-related domain, which is found primarily in Archaea. The archaeal promoters of acs4, filI, acs1, and mtrABC also initiated FilR1-modulated expression in an Escherichia coli lux reporter system, suggesting that FilR1 can up-regulate both archaeal and bacterial transcription. In conclusion, this work identifies an archaeal FilI/FilRs TCS that regulates the methanogenesis of M. harundinacea.


Introduction
Methanogenesis is a major contributor to global warming, and methanogens are the only organisms known to perform this metabolism [1].Acetate-derived methane contributes about 70% of the global methane production and is produced by acetoclastic methanogens such as Methanosarcina and Methanosaeta.Although these archaea possess prokaryotic cells, their genetic machinery for replication, transcription, and translation more closely resembles that of Eukarya than Bacteria [2].Methanogens are archaea that are distributed in diverse anoxic environments and can develop complex regulation mechanisms in response to environmental changes [3,4].
Two-component signal transduction systems (TCSs) are one of the principal means by which bacteria respond to environmental changes [5][6][7].Gao and Stock summarized the diverse bacterial TCSs, indicating that the typical TCS comprises a membranebound sensor histidine kinase (HK) and a cognate response regulator (RR) and catalyzes a phosphotransfer between the two [8].Typically, upon sensing environmental stimuli, the HK is autophosphorylated at a conserved histidine residue.The phosphoryl group is then transferred to its cognate RR at a conserved aspartate residue.In prototypical HKs, the cytoplasmic kinase core consists of a well-conserved C-terminal HK-like ATPase (HATPase) and a less-conserved histidine kinase A (HisKA) domain [9], which can be coupled to an overwhelmingly diverse array of sensory domains, including PAS, GAF, and HAMP domains, enabling HKs to sense a wide variety of stimuli.Characteristic RR superfamily proteins contain a receiver (REC) domain, which is linked to variable effector domains that mediate nucleic acid binding or enzymatic and protein/ligand-binding activity [8].Moreover, approximately 17% of RRs exist as standalone REC domains that lack an effector domain [8,10,11]; these RRs usually implement a more intricate regulation of a TCS [12,13].
TCS genes are found in many sequenced archaeal genomes; however, they are poorly studied [7,14,15].Methanosaeta harundinacea 6Ac is an obligate acetoclastic methanogenic archaeon that was isolated from an upflow anaerobic sludge blanket reactor [16].It contributes to granule formation and the efficacy of waste removal in the reactor.Previously, we determined that FilI, a bacterial LuxI homolog from M. harundinacea, synthesized the quorum sensing (QS) signal molecules carboxyl-acyl homoserine lactones (AHLs), which regulate the cell morphological transition from short cells to filaments and carbon metabolic flux from biomass formation to methane production [17].
In this work, based on in silico analysis, filI, the cotranscribed gene filR2, and the adjacent gene that we tentatively named filR1 were predicted to encode the only TCS in this archaeon.Then, through a combination of in vivo and in vitro approaches, FilI and the two FilRs were shown to possess many of the properties of bacterial TCSs.Moreover, they were found to be involved in the regulation of methanogenesis in M. harundinacea.

In silico analysis predicted a putative TCS in M. harundinacea
To gain a general view of the TCSs in M. harundinacea 6Ac, bacterial prototypical kinase core domains of HKs and REC core domains of RRs were used as probes to query the genome (CP003117) [17].The search found only three putative HKs (Mhar_0446, Mhar_0936, and Mhar_1766) with characteristic cytoplasmic HisKA and HATPase domains, and it identified five RRs (Mhar_0169, Mhar_0445, Mhar_0447, Mhar_1520, and Mhar_2042) with REC core domains.These results were consistent with those of the P2CS and MiST2 databases [18,19].Further, transmembrane regions of the three putative HKs were analyzed using three programs: TMHMM (http://www.cbs.dtu.dk/services/TMHMM/) [20], TMpred (http://www.ch.embnet.org/software/TMPRED_form.html), and SOSUI (http://bp.nuap.nagoya-u.ac.jp/sosui/sosui_submit.html) [21].All three programs revealed that only Mhar_0446 was likely to possess the two transmembrane regions that are common in prototypical membrane-bound HKs (Figure 1A, 1B).Next, functional domains of the five RRs were analyzed using the Pfam database (http://pfam.sanger.ac.uk/) [9].Mhar_0447 and Mhar_1520 were stand-alone RRs that possessed only the REC domain (Figure 1C), while Mhar_0169 and Mhar_2042 had REC domains that were fused with PAS sensory domains, which is an unusual RR organization in which the REC domain is fused to a ligand-binding domain [10].Only Mhar_0445 appeared to encode a typical TCS RR with a REC core domain coupled with PB005323, a predicted ArsR-related domain with potential DNA-binding ability (Figure 1C).Moreover, Mhar_0446 and Mhar_0447 could constitute an operon because they are only separated by a short intergenic distance of 37 bp and are adjacent to Mhar_0445.Therefore, the three proteins might constitute the only typical TCS in M. harundinacea.
Mhar_0446 encodes a protein of 886 amino acid residues that exhibits 39.6% identity to AhlI, a LuxI family autoinducer synthase of Erwinia chrysanthemi [17].The recombinant His 6 -tagged protein had carboxyl-AHL synthase activity; therefore, Mhar_0446 was named FilI [17].Using the Pfam database, six functional domains were predicted in FilI, including a CHASE4 domain, a HAMP (potential signal transmission) domain, a PAS intracellular stimuli sensory domain, a GAF intracellular stimuli sensory domain, a C-terminal HisKA domain, and an HATPase domain (Figure 1A, 1B).Moreover, the transmembrane region prediction programs TMHMM, TMpred, and SOSUI indicated that the CHASE4 domain and the HAMP domain are extracytoplasmic or transmembrane, while the other four domains are all cytoplasmic (Figure 1B).In addition, the AHL synthase activity of FilI suggested a potential connection between the predicted FilI/FilRs TCS and quorum sensing (QS) in M. harundinacea.
An alignment of the two RRs in this putative TCS (Figure 1C), named FilR1 and FilR2, indicated a sequence similarity of 48% between the C-terminal REC domains of FilR1 and FilR2 (Figure 1D).Moreover, according to the domain compositions, FilR1, with a potential DNA-binding domain, was predicted to be the direct effector of this putative TCS.FilR2, as a stand-alone RR, was predicted to be involved in yet-unknown regulation mechanisms.
filI and filR2 constituted an operon and were cotranscribed The in silico analysis predicted that filR2 and filI constitute an operon from which filR1 is divergently transcribed (Figure 2A).To test this prediction, a reverse transcription-polymerase chain reaction (RT-PCR) assay was performed using the total RNA of M. harundinacea as a template.As shown in Figure 2B, a PCR product was amplified from the cDNA of the filI-filR2 intergenic region, indicating the cotranscription of filI and filR2 (Figure 2B).As a control, no amplification was produced from the intergenic region between filR2 and the downstream ferredoxin gene (Figure 2C).

Autophosphorylation of FilI and phosphotransfer among FilI and the FilRs
A functional TCS relies on the autophosphorylation of the HK and a phosphotransfer from the HK to the cognate RR [8].To determine if these activities are present in the predicted FilI/FilRs TCS system, recombinant FilI, FilR1, and FilR2 were purified from Escherichia coli (Figure 3A).FilI was readily autophosphorylated within 45 min of incubation with [c-32 P] ATP (Figure 3B).Phosphorylated FilI was capable of robust phosphoryl group transfers to both FilR1 and FilR2 within a short time (Figure 3B).Autophosphorylation of the truncated FilI protein, containing only the C-terminal kinase core domain, was also detected.This domain was able to phosphorylate FilR2, but its efficiency was lower than that of the entire FilI protein (Figure 3C).In contrast, the truncated FilI only weakly phosphorylated the other two potential RR transcriptional regulators Mhar_0169 and Mhar_1520 after 30 min (Figure 3D).These results demonstrated the specificity of the phosphorelay from FilI to FilR2 and to FilR1, and demonstrated the activity of the FilI HK domain.Further, the FilI-synthesized carboxyl-AHLs (N-carboxyl-C10-HSL, m/z 318, or N-carboxyl-C12-HSL, m/z 346) were added directly to the autophosphorylation reaction of FilI and phosphortransfer reactions to FilRs.However, very weak enhancing effects were observed (Figure S1).Therefore, the FilI-synthesized carboxyl-AHLs did not have an evident direct effect on the autophosphorylation of FilI or phosphotransfer to FilRs in the in vitro experiment.
Chromatin immunoprecipitation-quantitative PCR (ChIP-qPCR) assay of the in vivo associations of FilR1 with the promoters of the TCS system and the key genes of methanogenesis Because of the lack of a genetic system for M. hurandinacea 6Ac, the in vivo TCS action of FilI and FilR1 was determined using ChIP assays.M. harundinacea 6Ac possesses a thick cell envelope; therefore, a novel method to gently lyse the cells with dithiothreitol (DTT) treatment at alkaline pH was developed from a procedure for Methanospirillum [22].As shown in Figure 4A, the promoter regions of filR1 and the filI-filR2 operon were enriched in the anti-FilR1 antibody-immunoprecipitated DNA (lane AbFilR1) but not in the mock-IP control DNA (lane CK).Moreover, compared with the mock-IP DNA (CK), qPCR indicated 11.7-and 24.8-fold enrichments for the promoters of filR1 and the filI-filR2 operon in the anti-FilR1 antibodyimmunoprecipitated DNA (Figure 4B), respectively.In the same experiment, there was no enrichment for a fragment of similar length from the 16S rRNA gene (Figure 4).Therefore, ChIP-PCR and ChIP-qPCR assays verified the in vivo association of FilR1 with its promoter and the promoter of the filI-filR2 operon, a common characteristic of TCSs.
Previously, FilI-synthesized AHLs were found to regulate a transition of the cellular morphology from short to long cells and a carbon metabolic flux leading to the formation of more methane and less cellular biomass from acetate in M. harundinacea [17].To test whether the FilI cognate RR FilR1 was involved in the regulation of methanogenesis, associations of FilR1 with the key genes in acetoclastic methanogenesis were assayed using the same protocol.As shown in Figure 5, the promoter regions of the acs1 operon, acs4 gene, mtr operon, fwdCABD operon, and omp gene (Table S1) were all amplified by PCR assays from the anti-FilR1 antibody-immunoprecipitated DNA samples (Figure 5A).Also, qPCR indicated 11-and 13-fold enrichments of the promoter regions of the acs1 operon and the acs4 gene, respectively, and modest enrichment of the mtr and fwd operons and the omp gene (6.34-to7.88-fold enrichment) (Figure 5B).Therefore, FilR1 interacted with the promoters of a number of genes that are essential for methanogenesis in M. harundinacea 6Ac as summarized in Table S1.FilR1 bound in vitro to the promoters of the TCS system and key genes of methanogenesis Next, electrophoretic mobility shift assays (EMSAs) were performed to confirm the in vitro binding of FilR1 to the gene promoters studied above.Using a biotin-labeled DNA amplified from the tested gene promoters as probes, DNA-protein complexes with the recombinant FilR1 were detected by native polyacrylamide gel electrophoresis (PAGE).As shown in Figure 6, FilR1 bound efficiently to its promoter and that of the filI-filR2 operon in a dose-dependent manner, demonstrating its auto-regulation and regulation of the filI-filR2 operon.
Similarly, EMSA also detected the in vitro binding of FilR1 to promoters of the acs1 operon, acs4 gene, mtr operon, fwdCABD operon, and omp gene (Figure 7).However, no binding was detected for the promoter of Mhar_0449 and an internal fragment of filI, which served as negative controls (Figure 7 and data not shown).These data further confirmed the direct regulation acetoclastic methanogenesis and the methyl oxidative shunt by FilR1; these findings were consistent with the in vivo binding results determined by the ChIP assays described above.In addition, the phosphorylated FilR1 via phosphotransfer from FilI exhibited an obviously enhanced DNA-binding affinity to the promoters of the acs1 and fwdCABD operons (Figure S2), implying the possible positive regulation of methanogenesis by the FilI/FilRs TCS in Methanosaeta.
However, EMSA did not detect the binding of FilR1 to the promoters of some other differentially expressed genes in the QSmediated transition.These genes included the cdhCD and cdhBEA operons, which encode the key enzyme complex for the acetoclastic reaction; the mcr operon, which encodes the methyl-CoM reductase; and the fpo operon, which encodes the only membrane-associated protein complex for electron transfer in Methanosaeta (data not shown).Therefore, the expression of these genes is probably not directly regulated by FilR1.
The ArsR-related domain is found primarily in Archaea, and it has not yet been shown to bind DNA.This work also provides the first experimental evidence of the DNA-binding ability of the ArsR-related domain.
An ex vivo E. coli promoter-reporter system detected FilR1 regulation of the genes involved in methanogenesis An E. coli-based promoter-reporter ex vivo system [23,24] was used to further examine the regulation of FilR1.First, the filR1 open reading frame (ORF) with a prototypical E. coli ribosomal binding site sequence was cloned into the expression plasmid p184 (abbreviation of pACYC184) to generate the plasmid pFilR1 (abbreviation of pACYC-FilR1) for heterogeneous expression of FilR1 protein in E. coli.Western blots with anti-FilR1 antibody detected a protein band of 54 kDa, the predicted size of FilR1, in extracts of transformed E. coli (data not shown).Meanwhile, the examined promoters were fused upstream of a promoterless lux operon inserted into pCS26-Pac to construct a variety of methanogenic promoter-lux reporter plasmids denoted as pO xlux (x: the tested genes) (Table 1).pFilR1 and each of the pO x -lux promoter-lux reporter plasmids were cotransformed into E. coli.p184 was cotransformed with the pO x -lux reporter plasmids as a negative control.As shown in Table 1, although the mtr, fwd, and cdhCD operons and the acs4 gene possessed archaeal promoters, they were recognized by E. coli RNA polymerase and promoted the transcription of the downstream lux gene, yielding luminescence comparable to that of bacterial promoters.Coexpression with FilR1 caused a remarkable stimulation of the acs4 (46-fold) and filI-filR2 (33-fold) promoters.FilR1 also increased the bioluminescence from the omp (13-fold), filR1 (8-fold), acs1 (6-fold), and mtr (3.7-fold) promoters (Table 1), suggesting that FilR1 regulated these genes as well.In contrast, the FilR1-stimulated expression by the promoter of the fwdCABD operon was increased less than 2-fold in this promoter-reporter system (Table 1).The fwdCABD promoter also possessed low affinity to FilR1 in EMSA (Figure 7D).Similarly, FilR1 did not affect the expression from the promoters of cdhCD, cdhBEA, mcrBD, mcrCA, fpo, and RNA pol, which are genes whose expression was not directly regulated by FilR.
To determine if archaeal promoters share the characteristics of bacterial promoters and, therefore, can be recognized by E. coli RNA polymerase, we first predicted the promoters in the abovestudied genes using the Neural Network Promoter Prediction program (http://www.fruitfly.org/seq_tools/promoter.html) [25] (Table S2).In addition to a typical archaeal promoter, a bacterialike promoter was found upstream of the mtr coding region (Figure S3).Coincidently, the highest bioluminescence was detected for the mtr and acs4 promoters in the E. coli promoter-reporter system (Table 1).Furthermore, a GC substitution was made for a six base-AT tract in the predicted promoters of mtr, acs4, filR1, and filI-filR.The bioluminescence from the mutated filI-filR2 and mtr promoters was reduced 6.4-and 31.5-fold,respectively, and FilR1-induced bioluminescence was also reduced (Table S3).The bioluminescence was not affected by mutations in the acs4 and filR1 promoters; this observation can probably be attributed to an incorrect promoter prediction.This indicates that E. coli RNA polymerase can recognize some archaeal promoters.

Discussion
TCSs are distributed widely in bacteria and are well characterized [7,14,19].They are abundant in free-living species like E. coli and Bacillus subtilis, which contain 30 and 36 TCSs [19], respectively, but they are absent in many parasites, such as Mycoplasma genitalium and Mycoplasma pneumonia [14,15].Because TCSs enable organisms to adapt to environments, an abundance of TCSs is considered a measure of the bacterial IQ [26].Authentic genes of archaeal TCSs are only found in some Euryarchaeota and Thaumarchaeota genomes, and they are found in neither Crenarchaeota nor Nanoarchaeota genomes [14,15].In these phyla, one-component systems of a single protein containing both sensor and regulator domains are widely distributed [27].Compared with bacteria, fewer TCSs are annotated in the sequenced archaeal genomes, and TCS HKs and RRs are not always present proportionally.Within Archaea, the most TCSs are found in Methanobacterium thermoautotrophicum (16 HKs and 10 RRs) and Archaeoglobus fulgidus (14 HKs and 11 RRs) [7].In contrast, Pyrococcus horikoshii contains a single HK and two RRs (chemotaxis proteins CheA, CheY, and CheB), and none are found in Methanococcus jannaschii, Aeropyrum pernix, and Thermoplasma acidophilum [7].
Hitherto, archaeal TCSs have not been well studied although this group of organisms is assumed to be highly adapted to extreme environments.In this work, an archaeal FilI/FilRs TCS, which is also the only in silico predicted TCS in the methanogenic archaeon M. harundinacea 6Ac, was identified experimentally, and the genes regulated directly by this TCS were determined.
TCSs are predicted to be of bacterial origin, and they were acquired by Archaea and some eukaryotes like plants, fungi, and protozoa by horizontal gene transfer [7,15].Though distinct from Figure 4. ChIP assays showed FilR1 binding to the promoters of its own (P filR1 ) and filI-filR2 operon (P filR1-filR2 ) inside the cells of M. hurandiacea 6AC. (A) PCR products of the promoters of its own (P filR1 ) and filI-filR2 operon (P filR1-filR2 ) were amplified from the anti-FilR1 antibody immunoprecipitated DNA (Ab FilR1 ), and input DNA sample before immunoprecipitation (Material and Methods) as a positive control (Input).Almost no PCR products were amplified from the mock-IP DNA (CK) samples.(B) qPCR detected the enrichment folds of the DNA fragments in anti-FilR1 antibody immunoprecipitated DNA (Ab FilR1 , gray bar) over mock-IP control (CK, black bar).PCR amplifications were performed using the specific primers for the promoter regions of filR1 (P filR1 ) and filI-filR2 operon (P filI-R2 ).An intragenic DNA fragment of the16S rRNA gene (16 s) was included as the negative control.doi:10.1371/journal.pone.0095502.g004 Figure 5. ChIP-PCR and ChIP-qPCR detected FilR1 associations with the promoters of genes for methanogenesis inside M. harundinacea cells. (A) PCR products amplified with the indicated primers using anti-FilR1 antibody immunoprecipitated DNA (AbFilR1) as a template and mock-IP DNA as a negative control (CK).(B) qPCR assays detected the enrichment of DNA fragments in anti-FilR1 antibody immunoprecipitated samples (AbFilR1, gray bar) over mock-IP negative control samples (CK, black bar).P acs1 , promoter region of operon acs1; P acs4 , promoter region of acs4; P mtr , promoter region of operon mtr; P fwd , promoter region of the operon fwdCABD; P omp , promoter region of omp and 16s, intragenic DNA fragment of 16S rDNA used as the control.doi:10.1371/journal.pone.0095502.g005 Bacteria, Methanosaeta 6Ac employs a TCS with properties that are similar to the bacterial system, i.e., catalyzing the phosphotransfer between FilI and FilRs, the HK and RRs of the TCS.Through ChIP-based in vivo assays and an EMSA-dependent in vitro approach, FilR1 was shown to bind the promoters of filI and its own gene, indicating that auto-regulation by the direct effector RR occurs in this archaeal TCS, a behavior that is typical of bacterial TCSs.FilR1 was also determined to bind to the promoters of genes involved in methanogenesis: the acs1 operon and acs4 gene encode acetyl-CoA synthetase for acetate activation at the initial and rate-limiting steps of acetoclastic methanogenesis; the mtr operon encodes a protein complex for methyl transfer from methyltetrahydromethanopterin to coenzyme M; and the fwdCAB operon encodes tungsten formylmethanofuran dehydrogenase subunits CABD in the methyl oxidative shunt.In addition, FilR1 was also determined to bind to the promoter of the omp gene, a cell envelope protein gene.These results indicate the direct regulation of acetoclastic methanogenesis and the methyl oxidative shunt by FilR1, which is inconsistent with the genes that are differentially expressed in QS-mediated filaments and short rod cells.Remarkably, a second RR protein, FilR2, is included in this archaeal TCS.This RR only has a REC domain and no output domain.The filR2 gene constitutes an operon with the gene encoding FilI, and the two genes are cotranscribed; therefore, an equal level of the two proteins might be maintained in cells.In addition, phosphoryl group transfer occurred from FilI not only to FilR1 but also to FilR2, implying the involvement of FilR2 in this phosphorelay.However, details of its function remain to be understood.Therefore, this work demonstrates that FilI, FilR1, and FilR2 constitute a methanogen TCS that is involved in the regulation of methanogenesis.
QS regulation has been extensively studied in Bacteria [28][29][30][31].Recently, a new signal transduction pathway in Vibrio harveyi was identified in which the QS signals, AHLs, are transduced via a TCS circuit [8,[31][32][33].However, QS regulation seems to be overlooked in Archaea.In our previous work, FilI, presumably the N-terminal fragment including the CHASE4 domain, was shown to catalyze the synthesis of carboxyl-AHLs, the QS signal of M. harundinacea that regulates the cellular transition from short cells to filaments.Physiologically, filamentous cells channel the substrate acetate more to methanogenesis than to biomass, implying a decrease in the methane yield [17]. Attempting to find the link between QS and the TCS in M. harundinacea, differential transcription in filaments and short cells was analyzed by RNA sequencing in a separate study.The analysis indicated that genes involved in acetoclastic methanogenesis (especially acs4), electron transfer reactions to generate reducing equivalents, ATP synthesis, and membrane-bound transporters were all upregulated in the filaments; these results were consistent with the phenotypes (unpublished data).The experimental data in this work suggests that FilR1 directly regulates acetoclastic methanogenesis through its interactions with the promoters of the key methanogenesis genes, supporting the relationship between the FilI/FilRs TCS and QS regulation in M. harundinacea.Although no evidence was found for direct effects of the FilI-synthesized carboxyl-AHLs in this work, including the effect of carboxyl-AHLs on the autophosphorylation of FilI and the phosphotransfer reaction between FilI and FilRs (Figure S1), a preliminary ChIP-PCR experiment showed that FilR2 association with the promoters of filR1, asc4, mtr, and omp occurred only in the filaments.This suggests that the stand-alone RR FilR2 can act as a candidate QS-related RR in this archaeal TCS, presumably through cooperation with FilR1 or other regulators.
This work determined that the Methanosaeta mtr, fwd, cdhCD, and acs4 promoters induced the transcription of a bacterial lux gene in E. coli.Because the expression was activated by the coexpression of FilR1, it mimics the natural expression in Methanosaeta and is unlikely to have resulted from fortuitous expression of cloned DNA near to but not including the promoter.Thus, some archaeal promoters appear to be recognized by bacterial transcription machinery, which is partially verified by mutation of the predicted promoters.This can also be anticipated on the basis of conserved promoter modularity and general basal transcription machinery among various lineages, as reviewed by Decker et al. [34].Recent biochemical and structural studies have revealed significant similarities of the basal transcription machinery among Bacteria, Archaea and Eukarya that include the following: 1) structural and functional similarity of the multi-subunit polymerases; 2) conserved promoter modularity from bacteria to humans; 3) functional equivalence between bacterial s factors and the basal transcription factors, TFB and TBP, in Archaea and Eukarya; and 4) similar strategies for transcription factors (TFs, activators, or repressors) to alter the core promoter specificity within Bacteria, Archaea, and Eukarya.While the detailed mechanisms for the action of the methanogen FilR1 with the E. coli transcription system need to be clarified through further studies, this work provides experimental evidence for the similarities in transcriptional machinery and regulation mechanisms between Bacteria and Archaea.
In conclusion, an archaeal FilI/FilRs TCS present in a methanogenic archaeon, M. harundinacea, was identified, and its regulation of methanogenesis was determined in this work.Moreover, this work indicated a possible connection between the FilI/FilRs TCS and QS regulation in this methanogenic archaeon; Values are shown as relative light units and the average of at least three independent readings.b.Genes and operons shown in each reporter plasmid are listed in Table S1.c.
Fold difference of the luciferase activity are calculated from those determined for E. coli strain carrying plasmid pairs of FilR1-vacant p184 plus pO x -lux over that of the strain carrying pOx-lux alone (+p184/2), strain with FilR1 plus pO x -lux over that with pO x -lux alone (+pFilR1/2), and strain with FilR1 plus pO x -lux over that with FilR1vacant p184 plus pO x -lux (+pFilR1/+p184).doi:10.1371/journal.pone.0095502.t001however, a detailed circuit pathway and mechanisms remain topics for further studies.

Strains, plasmids, and primers
M. harundinacea 6Ac was preserved in our laboratory and routinely cultured in a pre-reduced basal medium containing 100 mM sodium acetate in anaerobic bottles sealed with butyl rubber stoppers as described previously [16,17].E. coli DH5a served as the host for cloning experiments, while E. coli BL21 (DE3) was used for overproduction of recombinant proteins.E. coli strains were cultured at 37uC in Luria-Bertani (LB) medium with shaking.Kanamycin (50 mg/mL) and chloramphenicol (34 mg/ mL) were supplied when necessary.
The expression vector pET28a was obtained from Novagen (Madison, WI, USA).Plasmids pCS26-pac and pACYC184 for the bioluminescence assay were kindly provided by Professor Keqian Yang (Institute of Microbiology, CAS, Beijing, China).The primers used for DNA amplification in this study (Tables S4, S5, S6) were synthesized by Sangon (Beijing, China).

Construction of expression plasmids
The ORFs of filR1 and filR2 were amplified from genomic DNA of M. harundinacea with the primer pairs P1/P2 and P3/P4 (Table S4), respectively.The PCR products were digested with NcoI/ EcoRI and NcoI/HindIII, and then they were cloned into the same sites of the vector pET28a, resulting in the expression plasmids p28FilR1 and p28FilR2, respectively.The p28FilI plasmid was used to express FilI as previously described [17].The DNA fragment encoding the C-terminal HK domain of FilI (amino acids 642-886) was amplified from genomic DNA with the primer pair P5/P6 (Table S4) and inserted into the NheI/HindIII sites of pET28a, resulting in the expression plasmid p28FilI-C.Two ORFs (Mhar_0169 and Mhar_1520) encoding proteins with predicted REC domains served as the negative control for the phosphorelay assay and were amplified with the primer pairs P7/ P8 and P9/P10 (Table S4) and cloned into plasmid pET28a to construct expression plasmids p28-0169 and p28-1520, respectively.PCR-amplified sequences were verified by DNA sequencing of all constructs.

Protein expression and purification
To overexpress and purify FilI, FilR1, FilR2, FilI-C, Mhar_0169, and Mhar_1520, E. coli BL21 (DE3) harboring each of the expression plasmids was cultivated in LB medium containing kanamycin to an optical density at 600 nm of 0.4 to 0.6 when isopropyl-D-thiogalactopyranoside was added at a final concentration of 0.1-0.5 mM.The induced cultures were allowed to grow for an additional 3 h.Cells were harvested by centrifugation, resuspended in lysis buffer containing 0.3 M NaCl, 20 mM imidazole, and 50 mM sodium phosphate buffer (NaH 2 PO 4 /Na 2 HPO 4 , pH 8.0), and lysed by ultrasonication.The lysates were centrifuged at 13,400 g for 30 min, and the His 6tagged recombinant proteins were purified from the supernatants with a Ni 2+ -nitrilotriacetic acid-agarose column (Novagen), followed by ion change chromatography through a Resource Q column (GE Healthcare, Pittsburgh, PA, USA) or by gel filtration chromatography through a Superdex 200 10/300 GL column (GE Healthcare) when further purification was needed.Proteins were concentrated with Amicon Ultrafree-15 concentrators (Millipore, Billerica, MA, USA) when necessary.Purified proteins were examined by sodium dodecyl sulfate (SDS)-PAGE, and protein concentrations were determined using a bicinchoninic acid (BCA) protein concentration assay kit (Pierce, Rockford, IL, USA) combined with a Bradford protein concentration assay kit (Pierce).

ChIP assay
M. harundinacea 6Ac was grown to the late exponential phase, corresponding to the high cell density mode of QS [17], and the cells were immediately fixed with 1% (v/v) formaldehyde for 10 min.Fixation was terminated by the addition of glycine to a final concentration of 125 mM.Culture (450 mL; ,5610 10 cells) was harvested and pelleted at 10,000 g.Cell pellets were washed twice with Na 2 CO 3 /NaHCO 3 buffer (50 mM Na 2 CO 3 / NaHCO 3 , 150 mM NaCl, pH 9.0), resuspended in an equal volume of Na 2 CO 3 /NaHCO 3 buffer containing 100 mM DTT, and incubated for 2 h in anaerobic bottles.Cells were then collected at 5000 g and stored frozen at 280uC.Upon thawing, cells were resuspended in lysis buffer (10 mM Tris-HCl, pH 7.5, 1% (v/v) Triton X-100, 0.05% (w/v) SDS, 1 mM phenylmethylsulfonyl fluoride (PMSF)) and incubated in an ice-bath for 30 min.The protein concentration in the supernatant was determined by the BCA protein concentration assay kit (Pierce).The protein in the supernatant was diluted to 3 mg/mL with immunoprecipitation (IP) buffer (10 mM Tris-HCl, pH 7.5, 1% (v/v) Triton X-100, 0.05% (w/v) SDS, 0.05% sodium deoxycholate, 200 mM NaCl, 1 mM PMSF), and sonicated using a Bioruptor UCD300 (Diagenode, Denville, NJ, USA) until the chromatin DNA was sheared to an average size of 200-500 bp (2-7.5 min/cycle, 30 s on/30 s off, high-power setting) according to the manufacturer's instructions.Cell debris was removed by centrifugation, and the supernatant was retained.
A 500-mL aliquot of the sample was used for each IP experiment, and at least 10 mL of each sample was reserved as an input control.Each 500-mL sample was first incubated with 20 mL Dynabeads Protein A/G beads (Life Technology, Carlsbad, CA, USA) for 1 h at 4uC and mixed on a gently rotating wheel.The beads were separated from the supernatant by binding DynaMag-2 (Life Technology) and discarded.Next, the supernatant was incubated with 50 mL anti-FilR1 rabbit polyclonal antibody, which was purified by antigen affinity chromatography, overnight at 4uC with gentle mixing.A parallel experiment without antibody was conducted as a negative control (mock-IP).The supernatant was then incubated with an additional 20 mL Dynabeads Protein A/G beads that were pre-incubated with 5 mg/mL bovine serum albumin for another 1 h at 4uC with gentle mixing.Then, immunoprecipitated complex-bound beads were separated as above and washed twice with IP buffer, twice with IP buffer containing 500 mM NaCl, and once with Tris-EDTA buffer (pH 7.5).
Immunoprecipitated complexes were eluted in 53 mL Stock Reverse Crosslinking Buffer (Life Technology), combined with 1 mL proteinase K and incubated at 55uC for 60 min to decrosslink.Control samples (10 mL) were treated with the same procedure.Samples were separated by DynaMag-2, and the beads were discarded.Supernatants and input control were then incubated at 65uC for 30 min to inactivate the proteinase K. Finally, the non-crosslinked DNA was purified using DNA Purification Magnetic Beads and the provided buffers (Life Technology) or by the QIAquick PCR Purification Kit (Qiagen, Dusseldorf, Germany).DNA concentrations were determined by the Quant-iT DNA Assay Kit (Life Technology), and DNA fragment sizes were checked by an Agilent 2100 Bioanalyzer.All ChIP assays were performed at least three times to ensure the reproducibility.

PCR and real-time qPCR assay
PCR and real-time qPCR were performed to determine the enrichment of FilR1-bound targets in the immunoprecipitated DNA samples.For PCR amplification, 1 mL input, IP, or mock-IP (CK) DNA samples were used in a 25-mL reaction mix containing 0.4 mM of each oligonucleotide primer.PCR amplification used rTaq DNA polymerase (Takara, Dalian, China) for 25 to 30 cycles.Then, 5-10 mL reaction mixture was analyzed by electrophoresis on a 1.5% agarose gel.For qPCR analysis, a 25-mL reaction mixture was prepared that contained 16SYBR Green Real-time PCR Master Mix (Toyobo, Tokyo, Japan), 0.4 mM each of forward and reverse primers (Table S4), and 1 mL input, IP, or mock-IP (CK) DNA samples as templates.The qPCR analysis was performed on a Mastercycler ep realplex real-time PCR machine (Eppendorf, Hamburg, Germany).Fold differences between samples were calculated as described previously [35].Briefly, the DCt value (normalized to the input samples) for each sample was calculated according to the equation: DCt [Ct (sample) 2 Ct (input)].Next, the DDCt was calculated by DDCt = DCt (IP sample) 2 DCt (mock-IP control).Finally, the fold difference between the IP sample and mock-IP control was calculated as 2 (2DDCt) .

EMSA
All the probes that were used were amplified with a biotinlabeled forward primer.Double-stranded DNA (dsDNA) probes targeting the intergenic regions upstream of the studied genes (Table S1) including their DNA promoter regions were amplified using the genomic DNA of M. harundinaceaas as the template (Table S5), while the ORF of filI and the predicted DNA promoter region of Mhar_0449 were amplified as control probes.The DNA probes generated via PCR amplification were purified by an agarose gel DNA Purification Kit (Qiagen).A standard EMSA was performed using a Light Shift Chemiluminescent EMSA Kit (Pierce) as recommended by the manufacturer with some modifications as follows.A standard binding reaction mixture (20 mL) contained 10 mM Tris-HCl (pH 8.0), 50 mM KCl, 10 mM MgCl 2 , 1 mM DTT, 2.5% glycerol, 1% Nonidet P 40, 0.5 mM EDTA, 1 mg/mL poly dI-dC, 10 fmol labeled DNA probe, and the indicated amount of purified His 6 -FilR1 used in each reaction.After incubating at 25uC for 20 min, samples were immediately loaded on a non-denaturing 5% polyacrylamide gel (with an acrylamide to bisacrylamide weight ratio of 80:1) in 0.56 Tris-borate-EDTA buffer and run at 150 V for 2.5 h.Then, the DNA-protein complex was transferred onto a nylon membrane and cross-linked using a GS Gene Linker UV Chamber (Bio-Rad, Hercules, CA, USA).The biotin-labeled DNA was detected by chemiluminescence.
To determine the effect of phosphorylation on FilR DNAbinding, FilI protein (5 pmol) was autophosphorylated by incubation with 50 pmol ATP for 45 min at 37uC.Then, autophosphorylated FilI was mixed with 0, 0.1, 0.5, 1, and 2.5 pmol FilR1 for 10 min.EMSA of the phosphorylated FilR1 protein was performed as above.
Construction of promoter-reporter strains for luciferase assays in E. coli E. coli-based promoter-reporter strains were constructed to detect the possible regulation of genes of interest by FilR1.Intergenic regions upstream of the tested genes including their promoters were amplified by PCR with primer pairs listed in Table S6 and genomic DNA of M. harundinacea as a template.The amplified DNA fragments were then ligated to BamHI/XhoIdigested pCS26-Pac to yield the reporter plasmids pO x -lux (x refers to the tested gene).To express the FilR1 protein in E. coli, the filR1 gene was amplified with the primer pair P37/P38 (Table S6) using the plasmid p28FilR as a template and introducing E. coli Shine-Dalgarno sequences and recognition sites for BamHI [23].After digestion with BamHI, the fragment was ligated to BamHI/ EcoRV-digested pACYC184 to yield the plasmid pFilR1.ATtract mutants were constructed by PCR with primer pairs listed in Table S6 and pO x -lux reporter plasmids as templates; PCR products were digested with DpnI to yield the pO x-mut -lux reporter plasmids.The bioluminescence of the E. coli reporter cultures was measured using a TD 20/20n single tube luminometer (Turner Biosystems, Sunnyvale, CA, USA) as described previously [23].All measurements were performed with duplicate samples, and all experiments were repeated at least three times.Table S2 Promoter searching for some genes of M. harundinacea 6Ac using Neural Network Promoter Prediction Program.(PDF)

Supporting Information
Table S3 Effect of mutation of predicted TATA box in archaeal promoters on their expression in the E. coli ex vivo reporter system.(PDF) Table S4 Primers used for construction of expression plasmids.

(PDF)
Table S5 Primers used for amplification of promoter regions of the tested genes.

(PDF)
Table S6 Primer pairs used for construction of reporter plasmids for luciferase assays.(PDF)

Figure 1 .
Figure 1.Schematic representation of the domain structures of FilI and FilR proteins.(A) Domain structure of FilI analyzed using programs of Pfam and NCBI blast.(B) Location analysis of domains in FilI through the programs of TMHMM, TMpred and SOSUI.(C) Domain structures of FilR1 and FilR2 were analyzed by programs of Pfam and NCBI blast.In addition, the amino acid identity (%) for the aligned fragments of FilR1 and FilR2 is shown on the right.(D) Protein sequence alignment of the REC domains of FilR1 (C-terminal 276-446aa) and FilR2 (the whole length) by software GeneDoc.Identical amino acids are shown with a black background, while similar amino acids are shown with a gray background.doi:10.1371/journal.pone.0095502.g001

Figure 2 .
Figure 2. Cotranscription of filI and filR2 in M. harundinacea.(A) Schematic arrangement of filI and fliR2 in the genome.(B) Agarose gel electrophoresis of PCR products amplified from the intergenic region between filI and filR2.(C) The intergenic spacer between filR2 and its upstream gene encoding a ferredoxin using the respective template labeled at the top of each gel: -, no DNA; RNA, total RNA extracted form M. harundinacea cells; cDNA, reverse transcripts from the total RNA; gDNA, genomic DNA of M. harundinacea.M, DL2000 marker with the sizes shown at the right.Primers spacer-IR2-R/F and spacer-gen-F/R were used for PCR reactions in (B) and (C), respectively.doi:10.1371/journal.pone.0095502.g002

Figure 3 .
Figure 3. Assays of the autophosphorylation of FilI and phosphotransfer from FilI to FilR1 and FilR2.(A) SDS-PAGE of the recombinant his-tagged FilI, FilR1 and FilR2 purified from E. coli.(B) An autoradiogram visualized the autophosphorylation of the recombinant entire FilI protein (2 mg) after incubation for 45 min at 37uC in the presence of [c-32 P]ATP (lane 2) and without FilI as the negative control (lane 1).Phosphotransfer was assayed by addition of 4 mg of His-tagged RRs into the autophosphorylation reaction of FilI for certain times before being visualized on SDS-PAGE.Lane 3 and 6, negative controls of the phosphotransfer reactions without FilI included.Lane 4 and 5, FilR1 phosphorylated for 2 and 5 min, respectively; lane 7 and 8, FilR2 phosphorylated for 2 and 5 min, respectively.Dotted arrows indicate nonspecific bands generated from [c-32 P]ATP.(C) An autoradiogram visualized the autophosphorylation of the N-terminal truncated FilI (FilI-C) and phosphotransfer assay under the same treatments as the entire protein.Lane 1, no truncated FilI included; lane 2, FilI-C incubated with [c-32 P]ATP for 45 min; lane 3, FilR2 incubated with [c-32 P]ATP for 10 min; lane 4, 5 and 6, FilR2 phosphorylated for 2, 5 and 10 min, respectively.(D) An autoradiogram visualized the autophosphorylation of FilI-C and phosphotransfer assay to other regulators under the same treatments as the entire protein.Lane 1, no truncated FilI included, lane 2, FilI-C incubated with [c-32 P]ATP for 45 min; lane 3 and 6, proteins Mhar_0169 and Mhar_1520 incubated with [c-32 P]ATP for 10 min, respectively; lane 4 and 5, Mhar_0169 phosphorylated for 10 and 30 min, respectively; lane 7 and 8, Mhar_1520 phosphorylated for 10 and 30 min, respectively.Solid arrows indicate the phosphorylated proteins: Pi-FilI, Pi-FilI-C, Pi-FilR1 and Pi-FilR2.doi:10.1371/journal.pone.0095502.g003

Figure 6 .Figure 7 .
Figure 6.EMSAs showed FilR1 binding to the promoters of its own and the operon filI-filR2.Purified recombinant FilR1 protein was incubated with 0.5 nM of biotin-labeled DNA in the standard binding reaction mixture at 25uC for 20 min, and then run on a native PAGE.Concentration of purified FilR1 protein was shown at the top of each lane.Unlabeled filI-filR2 promoter (NP filI-filR2 ) was used as a competitor substrate of FilR1, which was added at the final concentrations of 5, 25, 125 and 250 nM in lane 6 and 15, 7 and 16, 8 and 17, and 9 and 18, respectively.(A) P filI- R2 , promoter of the filI-filR2 operon; (B) P filR1 , promoter of filR1, and (C) U filI , an internal DNA fragment of gene filI.doi:10.1371/journal.pone.0095502.g006

Figure
Figure S1 Effect of FilI synthetic carboxyl-AHLs on the autophosphorylation of FilI and phosphotransfer from FilI to FilRs visualized on SDS-PAGE.(A) autophosphorylation of the recombinant FilI protein (2 mg) incubated with [c-32 P]ATP for 45 min at 37uC in the presence (+) or absence (2) of a carboxyl-AHL (N-carboxyl-C 10 -HSL,m/z 318,or N-carboxyl-C 12 -HSL, m/z 346 at final concentration 2 ng).(B) Phosphotransfer of the autophosphorylated FilI to His-tagged RRs (4 mg) for 5 min in the presence (+) or absence (2) of carboxyl-AHLs (Ncarboxyl-C10-HSL,m/z 318,or N-carboxyl-C12-HSL, m/z 346 at final concentration 2 ng).Autophosphorylation and phosphotransfer reactions without FilI included as negative controls.Solid arrows indicate the phosphorylated proteins: pi-FilI, pi-FilR1 and pi-FilR2.Dotted arrows indicate nonspecific bands.(TIF) Figure S2 EMSAs showed phosphorylation enhancing the DNA binding affinity of FilR1.FilI protein (5 pmol) was firstly autophosphorylated by incubation with ATP (50 pmol) for 45 min at 37uC, and then mixed with 0, 0.1, 0.5, 1, 2.5 pmol FilR1 for 10 min, respectively.The mixtures including phosphorylated FilR1 protein were incubated with 0.5 nM of biotin-labeled DNA in the standard binding reaction mixture at 25uCfor 20 min before electrophoresis on native PAGE.The final concentrations of the FilR1 proteins for EMSA were shown at the top of each lane.(A) P acs1 , promoter of the acs1 operon; (B) P fwd , promoter of the fwd operon.FilR1, purified recombination FilR1 protein; Pi-FilR1, FilR1 protein incubated with autophosphorylated FilI protein before EMSA.(TIF)

Table 1 .
Expression of the archaeal promoters in the E. coli ex vivo reporter system.