Establishing a Markerless Genetic Exchange System for Methanosarcina mazei Strain Gö1 for Constructing Chromosomal Mutants of Small RNA Genes

A markerless genetic exchange system was successfully established in Methanosarcina mazei strain Gö1 using the hpt gene coding for hypoxanthine phosphoribosyltransferase. First, a chromosomal deletion mutant of the hpt gene was generated conferring resistance to the purine analog 8-aza-2,6-diaminopurine (8-ADP). The nonreplicating allelic exchange vector (pRS345) carrying the pac-resistance cassette for direct selection of chromosomal integration, and the hpt gene for counterselection was introduced into this strain. By a pop-in and ultimately pop-out event of the plasmid from the chromosome, allelic exchange is enabled. Using this system, we successfully generated a M. mazei deletion mutant of the gene encoding the regulatory non-coding RNA sRNA154. Characterizing M. mazeiΔsRNA 154 under nitrogen limiting conditions demonstrated differential expression of at least three cytoplasmic proteins and reduced growth strongly arguing for a prominent role of sRNA154 in regulation of nitrogen fixation by posttranscriptional regulation.


Introduction
Methanosarcina mazei strain Gö1 belongs to the methylotrophic methanogenic Archaea and, due to its role in methane production, is of high ecological relevance [1]. It serves as an archaeal model for investigating nitrogen stress responses, salt adaptation, methane production from different substrates, energy metabolism, as well as analyzing the role of small RNAs as regulatory elements in stress responses [2][3][4][5][6][7]. Although it only grows under strictly anaerobic conditions, the organism is genetically tractable and single colonies can be obtained on agar plates, a general requirement for genetic studies [8,9]. However, genetic manipulation is restricted due to the fact that puromycin is the only selectable marker commercially available for methanoarchaea, which complicates generation of multiple mutations or even complementation experiments. Using Methanosarcina acetivorans, Metcalf and coworkers developed a so-called markerless exchange method using the hpt gene encoding hypoxanthine phosphoribosyltransferase as a counterselectable marker [10]. A Δhpt strain which shows resistance towards the toxic purine analog 8-aza-2,6diaminopurine (8-ADP) can be used for counterselection following integration of an nonreplicable plasmid containing the wild-type hpt gene and the desired mutation with flanking regions for recombination. The complete plasmid is integrated into the site of the desired mutation (pop-in) in the chromosome by a single homologous recombination event, making the strain sensitive to 8-ADP and allowing selection for puromycin resistance. The presence of 8-ADP permits selection for removal of the plasmid-based hpt gene (in concert with the vector backbone) by another single homologous recombination (pop-out) event. During this latter event, the gene of interest can be exchanged by the mutant construct [10]. Theoretically, allelic exchange takes place with a chance of 50% resulting in the desired mutant strain.
The goal of this study was to establish this method for M. mazei in order to allow markerless chromosomal deletion or point mutations of small regulatory RNA genes. To set up the system, a Δhpt strain as well as the allelic exchange Archaea vector containing the wild-type hpt gene for counterselection was generated. To validate the method, we deleted the small noncoding RNA sRNA 154 . This sRNA has been identified in a genome wide RNA-seq screen and shown to be differentially transcribed dependent on nitrogen availability [7]. We suggest that sRNA 154 plays a central role in nitrogen regulation in M. mazei, potentially adding another level of regulation to the known regulatory mechanism via the general nitrogen transcriptional repressor NrpR [11,12]. sRNA 154 is located in the intergenic region of MM3337 and MM3338 encoding a conserved and a hypothetical protein, respectively [7,13]. A potential NrpRI operator (GGTA-N6-TACC) has been identified in the promoter region of sRNA 154 gene implying that this small RNA is under direct control of the global nitrogen regulator NrpRI [7].

Bacterial Strains and Plasmids.
Strains and plasmids used in this study are listed in Table 1. Plasmid DNA was transformed into E. coli according to the method of Inoue et al. [14] and into M. mazei using liposome-mediated transformation as described recently [8,15].

2.2.
Growth. M. mazei wild-type and mutant strains were grown in minimal medium under a nitrogen gas atmosphere in 5 or 50 mL closed growth tubes, which were incubated at 37 • C without shaking [18,19]. To screen on 8-ADP, however, the concentration of yeast extract in the minimal medium was reduced from 2 g/L to 0.5 g/L. In general, the medium was supplemented with 150 mM methanol or 25 mM trimethylamine (TMA) and 40 mM acetate as carbon sources and reduced with 2 mM cystein and 1 mM sodium sulfide. For nitrogen limited growth, ammonium was omitted from the media; molecular nitrogen in the gas phase served as sole nitrogen source [19]. In general, the Methanosarcina cultures were supplemented with 100 μg/mL ampicillin to prevent bacterial contamination. For mutant selection, puromycin (5 μg/mL) was added to the medium, for counterselection during markerless exchange the medium was supplemented with 8-ADP (20 μg/mL). Growth was monitored by determining the optical density of the cultures at 600 nm (O.D. 600 ). M. mazei wild-type and mutant strains were grown on solid medium by carefully spreading the cells on 1.5% bottom agar containing 25 mM TMA as carbon source and incubated in an intrachamber incubator under a gas atmosphere consisting of 79.9% N 2 , 20% CO 2 , and 0.1% H 2 S. Mutants were selected by adding 5 μg/mL puromycin or 20 μg/mL 8-ADP (final concentration) to the agar. To identify positive pop-out mutants, single colonies derived in the presence of 8-ADP were streaked in parallel on plates complemented with puromycin and 8-ADP, respectively, to screen for puromycin sensitivity and 8-ADP resistance.  The PCR products obtained contained additional synthetic primer-mediated restriction sites which, for the 800 up stream product included a BamHI at the 5 end and EcoRI site at the 3 end and for the 800 downstream fragment an EcoRI site at the 5 end and KpnI site at the 3 end. Both fragments were restricted using BamHI/EcoRI and EcoRI/KpnI, respectively, and cloned into pBSK+ (Stratagene, La Jolla, Calif, USA) yielding plasmid pRS283. The allelic exchange vector for the markerless exchange was generated by amplifying the hpt gene from chromosomal M. mazei DNA using the primers Mm hpt for and Mm hpt rev with additional BamH1 and XhoI sites, respectively. The PCR fragment was digested using BamHI and XhoI and ligated to BamHI and XhoI linearized pBSK+ to generate plasmid pRS311. In order to provide the hpt gene of pRS311 with a strong archaeal promoter, the known pmcr promoter of Methanococcus voltae [9] was cloned upstream of the gene. This was achieved by amplifying pmcr with the primers pmcr BamHI and pmcr XhoI using pRS207 [8] as template. The PCR product was cloned into TOPO-TA-cloning vector pDRIVE (Qiagen, Hilden, Germany) yielding plasmid pRS269. Digestion of pRS269 with BamHI resulted in excision of the pmcr promoter that was cloned into the BamHI site located directly upstream of the hpt gene of plasmid pRS311, resulting in pRS320. Finally, the 1.7 kbp EcoRI fragment from pRS204 containing the pac-cassette under the control of the constitutive promoter (pmcr) and terminator (tmcr) from the mcr-gene of M. voltae was cloned into the unique NotI site of pRS320, generating plasmid pRS345. This plasmid was used for markerless allelic exchanges by cloning the desired mutation into its unique ApaI site. To construct the M. mazei sRNA 154 deletion mutant, approximately 1000 bp of the upstream-flanking region of the small RNA was amplified using the primer pair Mm s154 1 for and Mm s154 1 rev using genomic M. mazei DNA as template. The PCR product (937 bp) was digested with ApaI and XhoI (restriction sites provided by the primers; see underlined sequences) and ligated into the ApaI/XhoI-opened pMCL210 vector resulting in plasmid pRS606. A 1364 bp PCR product of the downstream region was generated by primers Mm s154 2 for and Mm s154 2 rev introducing an XhoI and a SmaI restriction site, respectively, which was subsequently cloned into an XhoI/SmaI-linearized pRS606, fusing the sRNA 154 flanking region together and thereby deleting sRNA 154 . The plasmid was designated pRS631. The complete deletion construct was excised using ApaI and SmaI, treated with Mung Bean nuclease, and ligated into pRS345, which was linearized with ApaI followed by a Mung Bean nuclease treatment yielding plasmid pRS632. All constructs were verified by sequence analysis.

PCR Analysis for Mutant Strain Confirmation.
Verification of sRNA 154 deletion was performed using the primer pair Mm s154 1 and Mm s154 seq rev A ∼250 bp product of the bla gene was amplified using the primers bla rev. and bla for. The primer pair pac1 and pac2 was used to generate a ∼300 bp product of the pac-cassette. Generally, 2 ng of chromosomal DNA was used as template.

RNA Preparation and Northern Blot Analysis.
Total RNA isolations and Northern blot analyses were performed essentially as described before [7], except that Isol-RNA Lysis Reagent (5 PRIME GmbH, Hamburg, Germany) was used for total RNA preparation.

Results and Discussion
As mentioned above, markerless exchange of alleles originally developed for M. acetivorans was applied for M. mazei using the hpt gene as counterselection marker [10] and successfully generated a null mutant of the gene encoding sRNA 154 . encodes a hypoxanthine phosphoribosyltransferase. Nevertheless, we decided to construct a Δhpt mutant by applying the markerless exchange method of Pritchett et al. [10] rather than screening for a naturally occurring hpt-deficient strain. A pKS bluescript derivative was constructed carrying 800 bp of both the 5 and 3 flanking chromosomal region of the M. mazei hpt gene fused together, thereby creating an hpt deletion construct (pRS283). The nonreplicating plasmid pRS283 was transformed into M. mazei * [8], which will be referred to as wild type, and successful integration into the chromosome via a single homologous recombination event was confirmed by the gain of puromycin resistance ( Figure S1A). Single colonies were then inoculated into liquid medium containing 20 μg/mL 8-ADP. Cells that carry the wild-type hpt gene on the chromosome are sensitive to 8-ADP unless hpt is obliterated by a pop-out event, removing both the hpt gene and the plasmid backbone ( Figure S1B). Unfortunately, the standard minimal medium used for M. mazei [18] cannot be used for this approach as 8-ADP had little effect on growth of the cells (Figure 1(a)). Yeast extract, which is presumably rich in purines and pyrimidines, might affect uptake of 8-ADP. Growth in media with significantly reduced yeast extract (0.5 g/L) clearly demonstrated that 20 μg/mL 8-ADP was inhibitory (Figure 1(b)). As expected, the M. mazei Δhpt grew in the presence of 8-ADP on standard and on yeast-reduced medium (Figures 1(a) and 1(b)). Single colonies of the M. mazei Δhpt mutant strain were obtained by plating on solid medium containing 8-ADP, which were subsequently tested for puromycin sensitivity and simultaneous 8-ADP resistance by streaking on the respective plates. To confirm deletion of hpt, colonies that showed the desired phenotype were subjected to Southern blot analysis (data not shown). In a second step, the allelic exchange vector pR345 was constructed containing the bla gene and pac-resistance cassette for selection in E. coli and M. mazei, respectively, as well as the hpt gene as counterselectable marker. To provide the hpt gene with a strong promoter, the native promoter was exchanged with promoter pmcr of M. voltae [9]. The unique ApaI site in pRS345 provided an insertion site for the mutant construct of interest.

Generation of a M. mazei ΔsRNA 154 Chromosomal
Mutant. To validate the system, we deleted the gene encoding the small RNA 154 which is transcribed exclusively under nitrogen limitation and supposedly plays a central role in nitrogen stress responses [7]. A deletion construct generated by fusing the flanking regions of sRNA 154 together was inserted into the ApaI site of the allelic exchange vector pRS345. The resulting plasmid (pRS632) was transformed into the M. mazei Δhpt strain followed by selection for popin/pop-out events as described above. Successful deletion of the sRNA 154 gene was evaluated by PCR. PCR verification  Figure 4). Cells were harvested in exponentially phase. Equal amounts of cell extracts (65 μg) were applied to a 12% SDS-PA gel, which was subsequently stained with Coomassie blue. Arrows 1-3 indicate protein bands with differential expression in M. mazei ΔsRNA 154 and wild-type strains. Marker: LMW marker GE healthcare.
with primers binding up-and downstream of sRNA 154 was performed yielding a PCR product of ∼1,100 bp for wild type and ∼940 bp for M. mazei ΔsRNA 154 . The PCR product representing the wild type was detected as expected in M. mazei wild type and the diploid strain with plasmid pRS632 inserted into the chromosome (Figure 2(a)). The respective amplicon for ΔsRNA 154 was clearly detected in the control (pRS632), in the diploid strain and was very prominent in all eight potential M. mazei ΔsRNA 154 mutants analyzed (Figure 2(b)). However, in seven out of the eight putative ΔsRNA 154 mutants, traces of PCR products corresponding to the product derived from sRNA 154 wild type were also observed. This might be explained by the fact that several archaea have been demonstrated to possess multiple genome copies, as has been recently described by Soppa and coworkers [21]. They showed that M. acetivorans contains up to 17 copies dependent on the growth phase [21]. This polyploidy might result in incomplete allelic exchange with some of the chromosome copies remaining wild type. Since we could only confirm one out of eight mutant candidates, it appears that this difficulty occurs more often than anticipated when generating chromosomal mutants of M. mazei.
The mutant depicted in lane 11 (Figure 2(b)) showing the ΔsRNA 154 PCR product was further examined for plasmid removal. PCR analyses using chromosomal DNA from M. mazei wild type, the Δhpt mutant, the diploid strain, and ΔsRNA 154 as well as pRS632 as positive control clearly demonstrated the presence of the bla and pac genes 6 Archaea exclusively in the diploid strain and the plasmid control, whereas for ΔsRNA 154 , both genes were not detectable. As a second line of evidence, Northern blot analyses were performed with total RNA derived from the wild type, Δhpt and ΔsRNA 154 strains grown under nitrogen limitation and using a radioactively labelled oligonucleotide probe against sRNA 154 . Consistent with the previous data, Northern blot analyses clearly demonstrated that sRNA 154 with a size of 130 nucleotides (nct) is present in the wild type and Δhpt strains under nitrogen limitation but is not detectable in the sRNA 154 deletion strain, further confirming successful markerless allelic exchange ( Figure 3). By generating this ΔhptΔsRNA 154 mutant, which will be referred to as ΔsRNA 154 strain, we have effectively established the markerless exchange system in M. mazei.  (Figure 4(a)). Nevertheless, ΔsRNA 154 did not reach the same final cell densities as the wild type. Negative effects on nitrogen fixation due to the absence of the hpt gene were excluded by analysing growth behaviour of the parental strain (M. mazei Δhpt) (Figure 4(a)). As expected, no different growth phenotype of these three M. mazei strains was observed under nitrogen sufficiency as under this condition the sRNA 154 is not transcribed (Figure 4(b)).

Characterization of the
Characterizing the protein expression patterns of ΔsRNA 154 under nitrogen limitation and nitrogen sufficiency by one-dimensional SDS-PAGE clearly demonstrated differences in the protein patterns only under nitrogen depletion ( Figure 5). At least three different proteins were differentially synthesized under nitrogen limitation in the absence of sRNA 154 in comparison to the wild type ( Figure 5(a)). Two proteins (1 and 2) with the molecular mass of approximately 66 and 40 kDa were exclusively or significantly more strongly expressed in the mutant, whereas a 35 kDa protein (3) was present in the wild type but appears to be absent in the mutant. These findings indicate that sRNA 154 controls the protein expression either directly or indirectly and again strongly support a prominent function of the sRNA 154 in nitrogen regulation.
The ΔsRNA 154 mutant represents the first chromosomal deletion mutant of a small RNA in M. mazei. As it is only transcribed under nitrogen fixing conditions, presumably under the control of the global nitrogen regulator NrpRI [7], we suggest that sRNA 154 plays a central role in regulation of nitrogen metabolism. The differences in the cytoplasmic protein patterns that result in reduced growth of ΔsRNA 154 under nitrogen fixing conditions argue for a prominent role of sRNA 154 in regulation of nitrogen fixation. Posttranscriptional regulation by sRNA 154 would add another level of regulation of nitrogen metabolism in M. mazei possibly resulting in tighter control or fine tuning of translation of the target mRNAs.

Conclusion
By generating a Δhpt strain and a plasmid for allelic replacements, we successfully applied the markerless exchange system to M. mazei. The method was further optimized by using medium with reduced yeast extract, thereby enhancing the toxic effect of 8-ADP during counterselection. Generation of ΔsRNA 154 revealed the role of sRNA 154 in nitrogen metabolism as demonstrated by reduced growth as well as differential synthesis of at least three proteins under nitrogen fixing conditions in the absence of sRNA 154 .