SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Methanothermobacter thermautotrophicus is a methanogenic Gram-positive microorganism with a cell wall consisting of pseudomurein. Currently, no information is available on extracellular pseudomurein biology and so far only two prophage pseudomurein autolysins, PeiW and PeiP, have been reported. In this paper we show that PeiW and PeiP contain two different N-terminal pseudomurein cell wall binding domains. This finding was used to identify a novel domain, PB007923, on the M. thermautotrophicus genome present in 10 predicted open reading frames. Three homologues were identified in the Methanosphaera stadtmanae genome. Binding studies of fusion constructs of three separate PB007923 domains to green fluorescent protein revealed that it also constituted a cell wall binding domain. Both prophage domains and the PB007923 domain bound to the cell walls of Methanothermobacter species and fluorescence microscopy showed a preference for the septal region. Domain specificities were revealed by binding studies with other pseudomurein-containing archaea. Localized binding was observed for M. stadtmanae and Methanobrevibacter species, while others stained evenly. The identification of the first pseudomurein cell wall binding domains reveals the dynamics of the pseudomurein cell wall and provides marker proteins to study the extracellular pseudomurein biology of M. thermautotrophicus and of other pseudomurein-containing archaea.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Methanothermobacter thermautotrophicus (formerly Methanobacterium thermoautotrophicum) is a thermophilic, methanogenic archaeon which grows chemolithoautotrophically on the conversion of hydrogen and carbon dioxide to methane. Next to a model organism for methanogenesis and archaea in general, M. thermautotrophicus has been investigated for the archaea-specific cell wall, named pseudomurein, pioneered by König and Kandler almost three decades ago (Kandler and König, 1978; 1998). As already suggested by the name, the pseudomurein cell wall resembles murein (or peptidoglycan) of eubacteria in structure and function. Both cell wall types are composed of oligomeric heterodisaccharides that are connected by short peptide linkers and both determine cell morphology and protect bacteria from their direct environment and internal osmotic pressure. Despite these similarities, clear differences exist between murein and pseudomurein. In contrast to murein, pseudomurein (small variations have been described), is composed of N-acetyltalosaminuronic acid and N-acetylgalactosamine (both instead of muramic acid) connected to N-acetylglucosamine through a β-1,3-glycosidic bond (instead of a β-1,4-glycosidic bond in murein) (König and Kandler, 1979a; König et al., 1983). The peptides that link parallel glycan strands are all in the l-anomeric form, whereas murein connects its glycan chains with amino acids exclusively in the d-configuration. Also the biosynthesis of murein and pseudomurein shows significant differences (Hartmann and König, 1990a). Murein precursor biosynthesis starts by the generation of two nucleotide activated N-acetylated glucosamine (UDP-GluNAc) molecules. One UDP-GluNAc is converted to N-acetylated muramic acid with the subsequent addition of amino acids resulting in an activated N-acetylmuramic acid pentapeptide. The N-acetylmuramic acid pentapeptide is finally transferred to undecaprenyl to which the second N-acetylated glucosamine is connected through a glycosidic bond. In contrast, the biosynthesis of pseudomurein already starts with an UDP-activated disaccharide (König et al., 1989). The biosynthesis of the peptide moiety begins with an UDP-activated glutamic acid to which l-amino acids are connected individually (König and Kandler, 1979b; Hartmann et al., 1990; Hartmann and König, 1994; König et al., 1994). The disaccharide and pentapeptide are finally combined in an UDP-activated precursor which is transferred to undecaprenyl (Hartmann and König, 1990b).

These small but fundamental differences in cell wall chemistry and biosynthesis have led to the hypothesis that murein and pseudomurein are unlikely to have evolved from a common ancestor and must be the result of convergent evolution (Hartmann and König, 1990a). This is supported by the phylogenetic distribution of murein and pseudomurein-containing microorganisms. Murein is widely distributed across the eubacteria, while pseudomurein so far has been identified only in the order of the Methanobacteriales and the closely related Methanopyrus kandleri (Kurr et al., 1991; Kandler, 1994). The convergent evolution hypothesis of murein and pseudomurein is supported by the results of genome sequencing projects. In the last decade, extensive data have been published on the enzymes and the encoding genes that are involved in murein biosynthesis and assembly (Scheffers and Pinho, 2005). No gene homologues or similar pathways have been identified in the genomes of microorganisms that contain a pseudomurein cell wall, which implies a completely different set of enzymes in these microorganisms involved in pseudomurein biology.

Presently, only two enzymes, PeiW and PeiP, originating from M. wolfeii and M. marburgensis prophages ΨM2 (Pfister et al., 1998) and ΨM100 (Stax et al., 1992; Luo et al., 2001) respectively, are known to hydrolyse pseudomurein and function as autolysins. Both enzymes contain a cysteine protease catalytic domain that cuts the linker peptide of pseudomurein, thereby removing the cell wall and lysing targeted cells (Luo et al., 2002).

Initial experiments at our laboratory suggested that PeiW not only lyses but also associates with M. thermautotrophicus cells. In this paper we present evidence that PeiW and PeiP contain different N-terminal pseudomurein cell wall binding domains (CBDs). These results were used to identify a novel pseudomurein CBD in the M. thermautotrophicus genome which was confirmed by binding studies. The identification of the first pseudomurein CBDs with homologues in the Methanosphaera stadtmanae genome, provides marker proteins to study the extracellular pseudomurein biology of M. thermautotrophicus and other species.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

The PeiW and PeiP proteins are the only enzymes that are known to degrade the pseudomurein cell walls of Methanothermobacter species specifically (Luo et al., 2002). Both proteins originate from Methanothermobacter wolfeii prophage ΨM100 and M. marburgensis prophage ΨM2 respectively, where they function as autolysins. The ΨM100 phage initiates its lysogenic state when M. wolfeii has been deprived of its energy source, hydrogen (Luo et al., 2001). Upon PeiW expression, M. wolfeii cells are lysed and the resulting cell free extract can be used to lyse other Methanothermobacter species (Kiener et al., 1987). Isolation and sequence analysis of the encoding gene has shown that the lysing characteristics of PeiW are most likely confined to the C-terminal halve (Luo et al., 2002). This part of the protein contains an endoisopeptidase domain, which shows homology with animal transglutaminases (Makarova et al., 1999) (Fig. 1). The catalytic triad of this protease centres around a cysteine residue, making this protein highly sensitive to oxidative agents. The PeiW N-terminus starts with a unique 40-amino-acid stretch followed by three 30-amino-acid repeats and no function has been attributed to this part of the PeiW protein. Similarly, the PeiP protein sequence contains an endoisopeptidase domain at the C-terminal end and an unrelated N-terminus composed of repeats of unknown function.

image

Figure 1. Schematic representation of the molecular architectures of the wild-type (WT) PeiW (top) and PeiP (bottom) proteins and their modified versions. The black boxes represent His-tags. At the right, the cell wall binding activity and lysing activity is indicated qualitatively. ND, not determined.

Download figure to PowerPoint

Initial experiments in our laboratory to prepare genomic DNA and cell free extracts of M. thermautotrophicus, using the PeiW protein indicated that the PeiW protein largely disappeared from the supernatant if cells were incubated with PeiW shortly and centrifuged immediately. These results suggested that the PeiW protein not only interacted catalytically but also bound the M. thermautotrophicus cell wall.

The heterologous PeiW protein associates specifically with M. thermautotrophicus cells

To explore the possibility of the PeiW protein interaction with M. thermautotrophicus cells further, the PeiW protein was mixed with M. thermautotrophicus cells, incubated, spun down and the resulting supernatant was analysed for unbound PeiW protein (Fig. S1). Titration of cells in standard culture medium or phosphate buffer with a fixed number of cells and increasing amounts of PeiW showed that 2–4 × 107 PeiW proteins bound to a single M. thermautotrophicus cell at the conditions tested. Furthermore, binding at room temperature was fast and approximately 90% of the added amount of PeiW disappeared from the supernatant within 1 min (Fig. S2). Similar results were obtained at 37°C and 60°C but lysis of M. thermautotrophicus did not allow following binding kinetics quantitatively. To test binding specificity of PeiW, control experiments were performed with the bacteria Escherichia coli and Lactococcus lactis that were regarded as representatives for microorganisms containing a Gram-negative and a Gram-positive peptidoglycan cell wall respectively (Fig. 2 and Fig. S3). Aspecific binding was observed to some extent with E. coli cells. Binding, however, was not detectable when NaCl was included in the assays at a concentration of 50 mM. Under this condition identical amounts of PeiW bound to M. thermautotrophicus cells and no binding was observed for E. coli and L. lactis cells.

image

Figure 2. Binding specificity of the PeiW-CBD/GFP fusion protein. The figure shows a fluorescence microscopic comparison of M. thermautotrophicus cells incubated with the fusion protein without pretreatment (A) and cells 10 min boiled in 5% (w/v) SDS (B). (C–E) show microscopic pictures of a mixture of M. thermautotrophicus (irregular long rods), E. coli (short rods) and coccoid L. lactis cells incubated with the fusion protein; C, phase contrast; D, green fluorescence; E, overlay of phase contrast and green fluorescence. The white bars represent 1 μm in (A–B) and 20 μm in (C–E).

Download figure to PowerPoint

The PeiP and PeiW protein have modular architectures and contain N-terminal CBDs

The PeiW protein was investigated further to pinpoint the M. thermautotrophicus cell wall binding capacity. One explanation for the observed results could be that the catalytic domain also binds the cell wall before hydrolysing it. To test this hypothesis experiments were performed with PeiW protein and cells in the presence of hydrogen peroxide. Addition of hydrogen peroxide to a final concentration of 0.15% already prevented lysis of M. thermautotrophicus cells completely but left cell wall binding activity unaffected. These results indicated that the binding capacity was not related to the lysing activity of PeiW and suggested that the PeiW cell wall binding property was not located in the C-terminal animal transglutaminase-like domain.

To locate the cell wall binding capacity of PeiW, two deletion constructs were tested for their ability to lyse M. thermautotrophicus cells and to bind the cell wall (Fig. 1). Incubations of a construct containing only the C-terminal domain of PeiW resembling animal transglutaminases was still able to lyse M. thermautotrophicus cells but did not show any detectable binding to M. thermautotrophicus cells upon SDS-PAGE analysis. The truncated version of the PeiW protein was much more susceptible to heat denaturation. In contrast to the complete PeiW protein, it was completely inactivated after a15 min incubation at 65°C, which is the optimal growth temperature of the organism. A second construct containing only the N-terminal region of PeiW showed no detectable lysing activity, but bound rapidly to M. thermautotrophicus cells. These results implied that the PeiW protein is a modular enzyme composed of a C-terminal catalytic domain and an N-terminal CBD that, to a certain extent, have thermostabilizing interactions. The PeiW architecture was confirmed by testing a construct consisting of the PeiW N-terminus, C-terminally flanked by a His-tagged green fluorescent protein (GFP). Incubating cells with this construct followed by fluorescence microscopy showed that the PeiW N-terminus directed the GFP to the cell wall of M. thermautotrophicus (Fig. 2A). Titration experiments at room temperature showed that a single cell could harbour 0.9–1.4 × 107 proteins. This is slightly less, but in the same order of magnitude as the binding capacity of M.  thermautotrophicus cells for the complete PeiW protein.

Similar results were obtained with the PeiP protein (Fig. 1). A binding test of the PeiP N-terminal domain fused to GFP showed that also this N-terminus could direct GFP to the cell walls of Methanothermobacter species specifically. Apparently both prophages independently developed an autolysin that binds to cell walls through N-terminal domains that do not share any detectable sequence similarity.

To identify the ligand of the N-terminal PeiW CBD, M. thermautotrophicus cells were boiled for 10 min in a 5% SDS solution to remove any soluble cell wall components and lipids. After washing the cells with medium, staining patterns were observed identical to untreated cells, indicating that the CBD bound a SDS-insoluble portion of the cell wall, most likely pseudomurein (Fig. 2B). This was corroborated by the observation that the PeiW protein was released completely after solubilization of the M. thermautotrophicus cell wall. Furthermore, no fluorescence was observed with protoplasts that were incubated with the PeiW CBD fused to GFP.

blastp searches with the endoisopeptidase domain of PeiW reveal a putative pseudomurein binding domain on the M. thermautotrophicus and M. stadtmanae genomes

The identification of the N-terminal pseudomurein CBD in PeiW opened the possibility to identify other putative CBDs in the published genome of M. thermautotrophicus (Smith et al., 1997). Earlier blastp searches of the non-redundant NCBI database with the PeiW primary sequence already identified open reading frame (ORF) number MTH412 as the closest homologue present in the M. thermautotrophicus genome (Luo et al., 2002). A close inspection of the MTH412 primary sequence at the Pfam website to analyse the domain architecture showed the presence of the transglutaminase domain at the C-terminal end and two copies of an unknown 140-amino-acid PfamB domain PB007923 preceded by a signal peptide. The PB007923 domain family at the Pfam website comprises 12 sequences and include 10 M. thermautotrophicus-specific ORFs (as of 12th June 2006, listed in Table 1). The particular domain was also identified by the ProDom sequence analysis software, where it is named PD006249. Sequence comparisons of the members of the PB007923 domain (PD006249) showed that this sequence family is relatively heterogeneous (Table S1). The program blast 2 sequences showed that some members only have slight sequence similarities to individual sequences while having high homology to other members, taken together averaging around 50% sequence similarity. Especially MTH523 and MTH301 encoding a signal peptide and a single copy of the PB007923 domain showed high sequence homology to each other but considerably lower sequence identities towards the other sequence family members. None of the repeat sequences showed high sequence homology to all other sequence family members. This was also apparent from a sequence alignment of this family (Fig. 3). The highest sequence conservation appeared to reside in the N- and C-terminal ends of this family.

Table 1.  Overview of sequences that contain the PB007923 domain.
ORF numberGenBankGI numberUniProtSizeLoc. PB007923Genome organizationAuxiliary domains and location
  1. Sequences are given from left to right with their ORF number (MTH stands for M. thermautotrophicus ORF, Msp stands for M. stadtmanae ORF), GenBank number, GI number, UniProt identifier (if available), protein size in number of amino acids, the sequence stretch that corresponds with the PB007923 domain, the genomic organization and the presence and location of auxiliary domains. PKD domain, considered to be involved in extracellular protein–protein interactions or protein–carbohydrate interactions, CARDB domain, involved in cell adhesion, NosD, periplasmic copper binding domain, CASH domain, involved in carbohydrate binding and Trans_glut core, transglutaminase domain.

MTH75NP_27521815678103O2617983736–171MTH72MTH75PF00801 184–255 (PKD)
MTH87NP_27523015678115O2619084036–171MTH87MTH88PF00801 184–255 (PKD)
MTH301NP_27544415678329O2640117260–171MTH300MTH301None
MTH357NP_27550015678385O26457735447–582MonocistronicPF01841 652–709 (Trans_glut core)
MTH359NP_27550215678387O2645953459–199/231–364MonocistronicPF01841 449–506 (Trans_glut core)
MTH412NP_27555515678440O26512583117–256/267–407MonocistronicPF01841 498–555 (Trans_glut core)
MTH523NP_27566615678551O2662318554–180MTH519MTH523None
MTH716NP_27585915678743O2681217551602–1755MonocistronicPF00801 252–327/1355–1430 (PKD)
SM00722 357–805 (CASH)
PF05048 806–849 (NosD)
PF07705 955–1061 (CARDB)
MTH719NP_27586215678746O26815574424–574MonocistronicSM00722 56–203 (CASH)
PF05048 204–247 (NosD)
MTH795NP_27593415678817O2688640537–169MTH795-MTH796PF01841 319–377 (Trans_glut core)
Msp_0046YP_44710884488876None881579–713MonocistronicNone
Msp_0658YP_44769984489467None251109–239MonocistronicNone
Msp_0775YP_44781184489579None616313–448Msp_0775–Msp_0776PF01841 533–589 (Trans_glut core)
image

Figure 3. Alignment of the PB007923 family members. Similar residues are highlighted in colours to emphasize sequence similarities. Sequences are given with their ORF numbers of the M. thermautotrophicus (MTH) and M. stadtmanae (Msp) genomes. Numbers refer to the section of the original sequence used in the alignment.

Download figure to PowerPoint

The distribution of the PB007923 domain (PD006249) was further explored by blastp searches of the genomes of M. kandleri (Slesarev et al., 2002) and M. stadtmanae (Fricke et al., 2006), two archaea that are known to contain a pseudomurein cell wall. A blastp search with the complete MTH412 sequence resulted in the identification of three homologues in the M. kandleri genome that contained the transglutaminase C-terminally but lacked the PB007923 domain. In contrast, the same search in the M. stadtmanae genome revealed three ORFs with a C-terminal transglutaminase domain of which one ORF, Msp_0775, contained a single copy of the PB007923 domain. A blastp search with this domain of the M. stadtmanae genome generated two additional ORFs, Msp_0046 and Msp_0658, each containing a single copy of the PB007923 domain (Table 1).

Interestingly, all of the 10 PB007923-harbouring ORFs from M. thermautotrophicus except one (MTH716) contained a signal peptide. Furthermore, all ORFs, except MTH301 and MTH523, possessed auxiliary domains that are considered to function extracellularly (Table 1). These include polycystic kidney disease (PKD) domain PF00801 (MTH75, MTH87 and MTH716) considered to be involved in extracellular protein–protein interactions or protein–carbohydrate interactions, CARDB domain PF07705 (MTH716) involved in cell adhesion, periplasmic copper binding domain NosD PF05048 (MTH716 and MTH719), CASH domain SM00722 (MTH716 and MTH719) involved in carbohydrate-binding and the above-mentioned transglutaminase autolysin domain PF01841 (MTH357, MTH359, MTH412, MTH795 and Msp_0775). The PB007923-encoding ORFs were also analysed for their chromosomal organization. A previous analysis of the M. thermautotrophicus genome has shown that an upper threshold of 55 nucleotides for intergenic spacers of ORFs transcribed in the same direction identifies all previously characterized polycistronic operons of M. thermautotrophicus (J.T. Keltjens, unpublished). This threshold was used to determine if PB007923-encoding ORFs were part of polycistronic operons in M. thermautotrophicus and M. stadtmanae and if the sequence homology of co-transcribed ORFs could give clues on the function of the operon. Seven (MTH357, MTH359, MTH412, MTH716, MTH719, Msp_0046, Msp_0658) of the 13 ORFs did not fulfil the criteria and are most likely transcribed as a single gene. Most of the ORFs in polycistronic ORFs were transcribed with only one other gene, except for MTH75 (operon MTH72–MTH75) and MTH523 (operon MTH519–MTH523). No clear sequence homology was found for potentially co-transcribed genes except for MTH72, which contained multiple tetratricopeptide repeats (PF00515 and PF07719) known to be involved in protein–protein interactions and protein complex assembly and for MTH796 that contains a putative cyclase domain (PF04199). Although a function for these polycistronic operons cannot be deduced, it is clear that the PB007923-containing ORFs are not distributed randomly across the M. thermautotrophicus genome and that they appear to cluster.

Four of the identified ORFs contained a C-terminal transglutaminase domain next to one or two copies of the PB007923 domain in the N-terminus. This overall molecular architecture strongly resembles the domain distribution of PeiW, suggesting that the PB007923 domain (PD006249) might be involved in cell wall binding.

The PB007923 domain constitutes a novel pseudomurein CBD

As discussed above, the PB007923 domain is present in ORFs that are similar in molecular architecture to PeiW, a protein that binds the pseudomurein cell wall of M. thermautotrophicus. To test if the PB007923 domain constituted another pseudomurein CBD, fusion constructs of PB007923 to GFP were tested. Three ORFs were selected that differed with respect to the position and PB007923 copy number in the original sequence: MTH75, which contains a single domain at the N-terminal end, MH359, which contains two copies at the N-terminus, and MTH716, which contains a single copy at the C-terminal end. Primers were designed to encompass a small part of the flanking region to minimize the possibility that parts of the PB007923 domain were not included in the fusion protein. All three constructs as well as the GFP protein alone were expressed in E. coli strain Rosetta 2 and purified using the His-tag of the GFP protein; purity was confirmed by SDS-PAGE (> 90%, results not shown). An incubation of cells followed by fluorescence microscopic investigation showed that all PB007923-containing protein preparations directed GFP to the cell wall of M. thermautotrophicus (Table S2). No binding was observed for the GFP alone. Neither was fluorescence observed with a mix of microorganisms lacking pseudomurein cell walls. The results clearly showed that the PB007923 domain functions as an independent entity and constitutes a domain that confers protein-specific pseudomurein cell wall binding activity. Furthermore, M. thermautotrophicus staining patterns were identical to cells treated with the PeiW CBD fused to GFP, indicating that also the PB007923 domain binds the insoluble pseudomurein.

Binding of the prophage and PB007923 CBDs to different types of pseudomurein reveals binding specificities and pseudomurein heterogeneity

The domains were further characterized by studying their binding to different methanogenic species having a pseudomurein cell wall. Pseudomurein has been observed only within the order of the Methanobacteriales and the closely related M. kandleri. The Methanobacteriales consist of the thermophilic genera Methanothermobacter (Wasserfallen et al., 2000) and Methanothermus and the mesophilic genera Methanobacterium, Methanobrevibacter and Methanosphaera. Although a general model is available for pseudomurein, several archaea contain a cell wall with small pseudomurein modifications, different molar ratios of cell wall constituents (König et al., 1982) and differing cell morphologies. Methanobrevibacter ruminantium and M. stadtmanae (König, 1986) pseudomureins, for example, are composed of cross-linking peptides that contain threonine or serine in stead of alanine respectively. Others like M. kandleri and Methanobrevibacter smithii contain an ornithine modification of the cross-linking peptide. M. thermautotrophicus and M. marburgensis have differences in the glycan moiety of their respective pseudomureins (König et al., 1982).

To test the binding behaviour of the CBDs identified, a collection of Methanobacteriales was incubated with the GFP fusions and analysed by fluorescence microscopy: M. wolfeii, M. marburgensis, M. smithii, Methanobrevibacter arboriphilus, M. ruminantium, Methanobacterium formicicum and Methanothermus fervidus (Table S2). Microscopic investigations showed that all GFP constructs tested (CBDs of PeiP, PeiW, MTH75, MTH359, MTH716) with the Methanothermobacter species gave identical results (Fig. 4A). Cells showed green fluorescence distributed evenly, with a ring-shaped form of higher fluorescence intensity at the septal region of dividing cells. No other ring-shaped forms could be discerned here. To test if the CBD ligands were expressed in a growth phase-dependent manner, M. thermautotrophicus was cultured in a fed batch reactor and sampled at different growth stages, starting from early exponential cells with an optical density (OD) at 600 nm of 0.05 to late stationary cells with an OD at 600 nm of 2. Testing of these different growth-phase cells revealed no differences and therefore no clear growth-phase dependency of CBD ligand presentation could be concluded.

image

Figure 4. Microscopic pictures of prophage and M. thermautotrophicus CBDs reacting with pseudomurein-containing genera. Left column, overlay of phase contrast and green fluorescence; middle column, green fluorescence; right column, phase contrast. Pictures measure 7.5 μm by 10 μm. A. PeiW-CBD/GFP M. thermautotrophicus. B. PeiP-CBD/GFP M. stadtmanae. C. PeiP-CBD/GFP M. fervidus. D. MTH75-CBD/GFP M. kandleri. E. MTH359-CBD/GFP M. formicicum. F. PeiP-CBD/GFP M. arboriphilus. G. MTH716-CBD/GFP M. arboriphilus. H. PeiP-CBD/GFP M. smithii. I. MTH75-CBD/GFP M. smithii. J. MTH359-CBD/GFP M. ruminantium.

Download figure to PowerPoint

The binding behaviour of the CBD fusions to other Methanobacteriales showed that also M. stadtmanae cells bound all constructs, albeit not uniformly. The incubation of the PeiW CBD fusion resulted in a patchy staining pattern without a clear localization, while all other CBD fusions seemed to prefer the septal regions of M. stadtmanae (Fig. 4B). Binding to M. formicicum cell walls was only detectable with the CBD fusions of PeiW, PeiP and MTH359 and showed patchy or grainy staining patterns without any clear, specific localization (Fig. 4E). M. fervidus bound with CBD fusions of PeiW, PeiP and MTH75, of which the PeiW and MTH75 CBD constructs resulted in cells with evenly distributed fluorescence (Fig. 4C). M. kandleri cells were stained evenly with PeiW, PeiP, MTH75 and MTH359 (Fig. 4D). Remarkably, some individual cells gave clear fluorescence with the PeiW CBD construct at the septal and polar regions. Clear localized binding was observed when Methanobrevibacter species were tested. M. arboriphilus for example reacted only with the PeiP and MTH716 CBDs, which preferentially associated with the septal region (Fig. 4F). In case of the MTH716 CBD incubation, also binding to the poles was discernable (Fig. 4G). Similar results were obtained with Methanobrevibacter smithii. Here, PeiW, PeiP and MTH75 CBD fusion incubations not only showed localization at the septal region but also gave ring-shaped signals at the equatorial positions but none at the poles (Fig. 4H and I). In contrast, M. ruminantium only reacted with the MTH359 CBD construct which notably bound to the equatorial position of the cells (Fig. 4J).

Summarizing, the analysis of a collection of Methanobacteriales that differ slightly in pseudomurein cell wall chemistry reveals that many CBDs stain the same organism in a similar fashion, but patterns differ between different species (Table S2). The heterogeneity of staining patterns of the Methanobacteriales likely reflects differences in pseudomurein cell wall biology.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Methanothermobacter thermautotrophicus is a methanogenic archaeon that has served as a model organism for methanogenesis and archaea in general and the organism has been one of the first microorganisms found to contain a pseudomurein cell wall. The pseudomurein cell wall is very similar but not identical to the cell wall polymer murein of eubacteria. They resemble each other in their physiological role but show fundamental chemical differences: different derivatized sugar moieties are used and the amino acids that are part of the connecting peptides differ in anomeric configuration. Also the biosynthesis of both cell wall precursors shows fundamental differences, which has led to the hypothesis that pseudomurein and murein must have arisen as two independent evolutionary events (Hartmann and König, 1990a). This is supported by the phylogenetic relationships of murein and pseudomurein-containing microorganisms: while murein-containing eubacteria are disperse, pseudomurein-containing archaea are confined to the phylogenetically distinct Methanobacteriales and the closely related M. kandleri. Also genome sequencing projects have not revealed any similarity between genes shown to be involved in murein biosynthesis/assembly (Scheffers and Pinho, 2005) and genes in microorganisms containing a pseudomurein cell wall. In fact, only two prophage proteins PeiW and PeiP have been shown to function as autolysins for pseudomurein-containing archaea and comprise the only two enzymes proven to be involved specifically in extracellular pseudomurein biology (Luo et al., 2002).

PeiW and PeiP are cell wall binding prophage autolysins

In this paper we present evidence that the N-terminal repetitive domains of PeiW and PeiP constitute pseudomurein CBDs. Binding appears to be specific for pseudomurein and approximately 1 × 107 PeiW molecules can bind to a ‘single cell’. This number of ligands is approximately one order of magnitude higher than other characterized cell wall binding proteins (Loessner et al., 2002). This can largely be explained by the cell length of M. thermautotrophicus. The organism is approximately 10-fold longer than organisms like E. coli and L. lactis (Fig. 2). Moreover, M. thermautotrophicus grows as filaments of two up to eight rods comprising individual cells (averaged length, 15 μm) that are separated from each other by proper cell walls as well as longer cells harbouring multiple chromosomes (Majernik et al., 2005). The latter are notably present in the exponential growth phase, the number decreasing during the stationary phase. Interestingly, CBD/GFP constructs preferentially bind to the septal region permitting an easy identification of the individual cells.

Binding of phage autolysins to the cell wall of their host has been reported previously. Indeed, the PeiW type of domain organization, for example, has also been observed for the autolysins of phages A500 and A118 of Listeria monocytogenes; however, in contrast to PeiW, both the protease and the binding domains of the enzymes were essential for catalytic activity (Loessner et al., 2002). Binding of murein hydrolases from viruses may involve the use of eubacterial peptidoglycan binding domains. Presently, two generally occurring domains are recognized at the Pfam website, notably PG_binding_1 (> 1150 entries) and the lysM domain (formerly PG_binding_2, > 2600 entries). A murein hydrolase encoded by Streptomyces aureofaciens bacteriophage m1/6 (Farkasovska et al., 2003) exemplifies the former; viral murein hydrolases possessing the lysM domain seem to be more frequent.

The PB007923 domain and peptidoglycan binding domains

The observation of regular murein binding domains in phage autolysins bearing similarity to CBDs of the host organism opened the possibility to use the PeiW sequence to identify potential pseudomurein CBDs in the genome of M. thermautotrophicus and in the available genomes of other pseudomurein-containing archaea. This resulted in the identification of a candidate domain of approximately 120 amino acids, PB007923 recognized at the Pfam website, which is identical to ProDom family PD006249. This domain is present in 10 M. thermautotrophicus proteins and has three homologues in the M. stadtmanae genome (Fricke et al., 2006). In comparison, PG_binding_1 and lysM are considerably smaller, sizing ∼40 and ∼30 amino acids respectively. So far, only 17 species of bacteria have been shown to contain 10 or more lysM-encoding ORFs, whereas only 14 species harbor 10 or more ORFs with the PG_binding_1 domain. The relatively high number of 10 M. thermautotrophicus pseudomurein binding proteins, most containing an additional, distinct (catalytic) domain might reflect an active, but presently not understood extracellular biology. Surprisingly, no copies of the PB007923 domain or other conserved regions were detected in the autolysin homologues that were identified in the M. kandleri genome. Perhaps the extremely high growth temperature imposes a different kind of strategy to direct autolysins to the M. kandleri cell wall during cell growth.

The PB007923 domain is a pseudomurein binding domain

To determine the ligand of the novel CBDs, M. thermautotrophicus was subjected to a pretreatment of 10 min in a boiling solution of SDS to remove any soluble parts of the cell wall. Binding studies gave identical results with pretreated and untreated cells. These results indicated that the prophage CBDs and PB007923 domains recognize pseudomurein. The observation that the septal regions of Methanothermobacter species appeared to bind more pseudomurein binding domains suggests that they might recognize a ligand produced at these sites. One obvious candidate would be freshly prepared cell wall material that is not yet cross-linked to other newly produced peptides. In fact this is consistent across the collection of pseudomurein-containing genera tested. Not only for the accumulation of GFP at the septal region but also at the equatorial planes for some genera. For murein biosynthesis it has been shown that some eubacteria, next to a septal plane of murein synthesis, also produce murein at the equatorial planes to allow cell extension (Scheffers and Pinho, 2005). Perhaps the visualization of the equatorial planes for the Methanobrevibacter species reflects evidence for a different type of cell division in comparison with the other archaea tested.

One genus reacted with only one CBD: M. ruminantium. This microorganism contains a cell wall with cross-linking peptides that have alanine substituted by threonine. The substitution might be an interesting site to prevent cell wall binding for most of the pseudomurein binding domains. However, the pseudomurein cell wall of M. stadtmanae has a linker peptide sequence that has the alanine substituted with serine (König, 1986), an amino acid very similar to threonine. Therefore, the differences in cross-linking peptides do not reveal a unifying ligand for the binding of the PB007923 domain. Also the differences in pseudomurein glycans tested did not clearly affect cell wall binding. Methanothermobacter marburgensis, in comparison with M. thermautotrophicus, contains large amounts of N-acetyl-galactosamine (König et al., 1982), which did not result in a detectable decrease in ligand binding. The directional localization of CBDs recognizing murein has been investigated for the lysM domain. Here, distribution appeared to be regulated by the cell wall constituent lipoteichoic acid and lysM binding was prevented by the presence of this compound (Steen et al., 2003). This modification or other types of putatively regulatory covalent modifications of the pseudomurein cell wall have not been reported. The targeted binding of the domains analysed therefore is unlikely to be directed by pseudomurein modifications but apparently through a specific recognition of a localized ligand. In the case of M. fervidus and M. kandleri, these ligands could have been shielded by the proteinaceous layer that surrounds these cells (Kurr et al., 1991) causing the patchy staining patterns. For these cells the exact nature of the ligand remains to be determined as well.

Taken together, the identification of these CBDs provides marker proteins to study the extracellular pseudomurein biology, which will shed light on the dynamics and the unexpected heterogeneity of this class of archaeal cell walls.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Bacteria, plasmids and culture conditions

Methanothermobacter thermautotrophicus (DSM1053) was cultured in 120 ml serum flasks or as a fed batch (3 l) as described before (Pennings et al., 2000). M. kandleri (DSM6324), M. stadtmanae (DSM3091), M. fervidus (DSM2088), M. smithii (DSM861), M. ruminantium (DSM1093), M. arboriphilus (DSM1125) M. wolfeii (DSM2970) and M. marburgensis (DSM2133) were obtained from the Deutsche Sammlung von Microorganismen und Zellkulturen (DSMZ, Braunschweig, Germany) as actively growing cultures. E. coli K12, L. lactis, Serratia marcescens, Staphylococcus aureus, Streptococcus faecalis, Bacillus subtilis, Pseudomonas aeruginosa, Penicillium chrysogenum and Saccharomyces cerevisiae served as negative controls in binding experiments and are part of our laboratory's strain collection.

Escherichia coli Top10 cells (Invitrogen) were used as general plasmid host and E. coli strain Rosetta 2 (Novagen) was used as a host for heterologous expression. Both E. coli strains were cultured in Luria–Bertani (LB) medium supplemented with 25 μg ml−1 kanamycin (pET30a vector, Novagen) or 100 μg ml−1 ampicillin (pGEM T-Easy, Promega and pET 100 vector, Invitrogen) as selective antibiotics respectively. The vector pEGFP-C2 was used as a source for the gene, gfp, encoding GFP and was a kind gift of Dr H. Alwan (Department Cell biology, Radboud University Nijmegen, the Netherlands).

The pET28a expression vectors encoding the N-terminally His-tagged pseudomurein endoisopeptidase PeiW and PeiP from prophages of M. wolfeii and M. marburgensis, respectively, were a kind gift of Dr A. Wasserfallen (ETH, Zürich, Switzerland) and Dr K. Sandman (Department of Microbiology, Ohio State University, USA).

General DNA techniques and cloning procedures

Standard DNA manipulations were performed essentially according to Sambrook et al. (1989). Enzymes were purchased from MBI Fermentas and used according to the manufacturer's recommendations. Genomic DNA of M. thermautotrophicus was isolated essentially as described before (Kiener et al., 1987). In short, the pseudomurein cell wall was degraded using the heterologous PeiW protein in standard medium supplemented with 0.5 M sucrose. Protoplasts were centrifuged and the supernatant was removed. Next, protoplasts were resuspended in milliQ and vortexed to stimulate cell lysis and the resulting suspension was centrifuged. Genomic DNA in the supernatant was purified using the QIAEX II clean-up kit (Qiagen). All PCR reactions were performed with the proofreading polymerase pfu. PCR products were cleaned directly or after agarose gel electrophoresis using the QIAEX II clean up kit. Purified PCR products were ligated directly in pET 100 vector or after A-tailing with Taq polymerase into the pGEM T-Easy vector. All constructs were confirmed by sequencing.

Sequence analysis

blastp searches (McGinnis and Madden, 2004) were performed at the NCBI website (http://www.ncbi.nlm.nih.gov/blast) against the non-redundant database at default settings or at the M. thermautotrophicus section of the PEDANT website (http://pedant.gsf.de/) (Riley et al., 2005). Signal peptides were predicted using the SignalP software (http://www.cbs.dtu.dk/services/SignalP/) (Bendtsen et al., 2004) using the settings for Gram-positive microorganisms. Domain searches were performed at the Pfam website (http://www.sanger.ac.uk/Software/Pfam/search.shtml) (Bateman et al., 2004) or at the ProDom (Corpet et al., 2000) website (http://protein.toulouse.inra.fr/prodom/current/html/form.php) at default settings. Domain similarities were taken from a blast 2 sequences comparison (Tatusova and Madden, 1999) (http://www.ncbi.nlm.nih.gov/blast/bl2seq/wblast2.cgi) at the NCBI website at default settings but with filtering disabled. Similarity is the percentage of identical and similar amino acids in the resulting alignment. The reported expect values are calculated for a search against the complete NCBI non-redundant database. The Multalin (Corpet, 1988) sequence alignment of the PB007923 family members was performed with the amino acid sequences identified at the Pfam website. The alignment was performed at default settings (http://prodes.toulouse.inra.fr/multalin/multalin.html). Polycistronic operon organization was derived from the direction of the ORFs and size of the intergenic spacers within a putative polycistronic operon, which was set at 55 bp.

Cloning of PeiW N- and C-terminal deletion constructs

The PeiW protein (AF301375) consists of a 41-amino-acid N-terminus with no significant sequence homology to known proteins, followed by a repetitive region containing three repeats running from residues 42–142, and a C-terminal endoisopeptidase domain encompassing residues 143–284.

Expression constructs of PeiW lacking the N- or C-terminal domain were constructed using a N-terminally His-tagged PeiW expression construct in pET28a (PeiW–pET28a) as starting material. The C-terminally deleted PeiW construct lacking the predicted endoisopeptidase catalytic domain was constructed using an unique DraI restriction site located immediately after the N-terminal repetitive region located at nucleotide 426 of the coding sequence. A DraI/XhoI double digestion with subsequent removal of the XhoI 3′-overhang with Klenow fragment resulted in a construct encoding a double His-tagged N-terminal region of PeiW. The expression construct encoding the C-terminal animal transglutaminase domain was amplified by PCR (for primers sequences, see Table S3), gel-purified and ligated into pET 100, resulting in a N-terminally His-tagged PeiW derivative without repeats followed by the C-terminal endoisopeptidase domain.

Construction of His-tagged pseudomurein CBD–GFP protein fusions

An expression vector was constructed containing the gene for GFP. The gene encoding GFP was amplified from plasmid pEGFP-C2 with a forward primers containing a NcoI restriction site and the reverse primers containing a BamHI site (Table S3). The PCR product was cloned into the NcoI/BamHI sites of the multiple cloning site of pET30a, resulting in pMyGreenpET, encoding a N- and C-terminally (double) His-tagged GFP product.

The N-terminal pseudomurein CBD of PeiW was amplified from the original expression vector PeiW-pET28a and encoded the complete N-terminus running from residues 1–151 with primers PeiW CBD Fwd and PeiW CBD Rev (Table S3). The PCR product was cloned in-frame N-terminally (NdeI/BglII) relative to the GFP in pMyGreenpET.

The putative pseudomurein CBDs of PeiP, MTH75, MTH359 and MTH716 were amplified from the PeiP-pET28a expression vector and from genomic DNA of M. thermautotrophicus respectively (Table S3). The CBDs of M. thermautotrophicus genes were selected by the number and the relative positions of the CBDs in the original predicted proteins: MTH75, one N-terminal CBD; MTH359, two N-terminal CBDs and MTH716, one C-terminal CBD. Each CBD was cloned in-frame with the GFP in such a way that the relative position of the CBD in the original protein was maintained, resulting in PeiP CBD-GFP, MTH75 CBD-GFP, MTH359 CBDs-GFP and GFP-MTH716 CBD. PCR primers of CBDs were designed to exclude the predicted signal peptide and contained restriction sites for cloning into the NdeI and BglII sites of pMyGreenpET, except for the PeiP CBD that was cloned in the NdeI/NcoI restriction sites. CBDs cloned C-terminally were ligated into the EcoRI/SalI restriction sites.

Production, purification and storage of heterologous proteins

Expression constructs in pET28a or pMyGreenpET were transformed to chemically competent E. coli Rosetta 2 cells that contained the pRARE2 plasmid. This plasmid compensates for tRNAs that are rarely used by E. coli and is maintained by chloramphenicol-resistance selection. Transformants were grown on LB plates containing 25 μg ml−1 kanamycin, 34 μg ml−1 chloramphenicol and 0.5% (w/v) glucose. Colonies were transferred to 5 ml pre-warmed LB medium containing 25 μg ml−1 kanamycin, 34 μg ml−1 chloramphenicol and 0.5% (w/v) glucose and incubated for 3 h at 37°C and 250 rpm. Hereafter, cells were transferred to 200 ml pre-warmed LB medium in a 1 l erlenmeyer with 200 ml LB medium containing 2.5 μg ml−1 kanamycin, 3.4 μg ml−1 chloramphenicol and 0.5% (w/v) glucose and grown at 37°C and 250 rpm until the OD at 600 nm reached 0.6–0.9. Next, the culture medium was cooled to 30°C and isopropylthiogalactoside (IPTG) was added to a final concentration of 1 mM. The induction was continued for 3–4 h at 30°C and 250 rpm. Expression of the catalytic domain of PeiW in the pET 100 vector was performed as described above except that pET 100 plasmid selection required 100 μg ml−1 ampicillin in stead of kanamycin. Furthermore, induction with IPTG was started 30 min after adding ethanol to a final concentration of 4% (v/v).

After induction, cells were harvested by centrifugation (10 000 g, 4°C, 15 min) and washed once with 50 ml Native Lysis Buffer (50 mM NaPO4 buffer pH 8, 10 mM imidazole, 300 mM NaCl) and resuspended in 25 ml Native Lysis Buffer. The cells were lysed by passing the cell suspension once through a French Pressure Cell operated at 138 MPa and cell free extract was obtained as the supernatant after centrifugation (20 000 g, 4°C, 20 min). Cell free extract was mixed with 10 ml bed volume of Ni-NTA resin (Qiagen) and incubated for 30 min on a rotary shaker (30 rpm) at room temperature. The resulting suspension was loaded onto a glass column and the resin was allowed to settle. The column with 10 ml Ni-NTA resin bedvolume was equilibrated in 50 ml of Native Lysis Buffer and washed with 100 ml Native Wash Buffer (50 mM NaPO4, pH 8, 20 mM imidazole, 300 mM NaCl) at a flow rate of 5 ml min−1. Bound protein was eluted with Native Elution Buffer (50 mM NaPO4, pH 8, 250 mM imidazole, 300 mM NaCl) at a flow rate of 0.5 ml min−1, collecting 1 ml fractions.

Protein concentrations were determined with the Protein assay solution (Bio-Rad) using bovine serum albumin as a standard. Fractions containing a protein concentration more than half the protein concentration of the peak fraction were pooled, mixed 1:1 with 87% glycerol (v/v) and stored, anaerobically at −20°C. The purified N-truncated PeiW protein was aliquoted, frozen in liquid nitrogen and stored at −80°C.

The homogeneity of purified heterologous protein preparations was confirmed by SDS-PAGE followed by Coomassie brilliant blue staining; purity was 90% or better.

Binding and activity measurements of complete and deletion constructs of PeiW to M. thermautotrophicus

Methanothermobacter thermautotrophicus cells (30 ml culture) were harvested by centrifugation (10 000 g, room temperature, 10 min) and washed once with 1 vol. of fresh culture medium. Cell pellets were resuspended in fresh medium to get the appropriate cell density. Binding and activity studies in the presence of hydrogen peroxide were performed in 50 mM sodium phosphate buffer, pH 7, 50 mM NaCl.

To determine the number of ligands per M. thermautotrophicus cell, increasing amounts of the protein to be tested were mixed in a series of vials, each containing an equal volume and number of cells. Cell numbers were derived from the OD measurement of the culture at which an OD600 = 1 equals 108 cells ml−1 (Kiener et al., 1987; own unpublished results). After incubation at room temperature, 37°C or 65°C for 10 min, the suspensions were centrifuged (10 000 g, room temperature, 2 min) and the supernatants were analysed by SDS-PAGE. It was assumed that binding ligands were fully saturated at the first point were added protein could be seen on the gel. The total number of ligands was estimated between this value and the highest amount applied without detectable protein in the supernatant.

PeiW activity measurements were performed essentially as described before (Luo et al., 2002). Washed cells were incubated at the specified temperature with PeiW or a truncated derivative and OD was monitored at 600 nm. Enzyme activity was deduced from a decrease of OD of the cell suspension.

Microscopic investigations of cells incubated with pseudomurein binding domains fused to GFP

A collection of pseudomurein-containing bacteria was ordered from DSMZ as actively growing cultures or cultured from our laboratory's strain collection: M. thermautotrophicus, M. wolfeii, M. marburgensis, M. formicicum, M. fervidus, M. stadtmanae, M. smithii, M. arboriphilus and M. ruminantium. A collection of Gram-positive and Gram-negative bacteria and two fungi were included as negative controls: E. coli K12, L. lactis, Serratia marcescens, Staphylococcus aureus, Streptococcus faecalis, Bacillus subtilis, Pseudomonas aeruginosa, Penicillium chrysogenum and Saccharomyces cerevisiae. Cells in 1 ml were collected by centrifugation (10 000 g, room temperature, 2 min) and washed once with 1 vol. fresh M. thermautotrophicus medium. After resuspension in 100 μl medium, cells were mixed with the construct to be tested and incubated for 10 min at 37°C. Next, cells were washed once with 1 ml culture medium and resuspended in 100 μl medium. Finally, an aliquot was transferred to a slide and analysed by phase contrast and for green fluorescence.

To rule out the possibility that GFP interacted in an aspecific way, cells were incubated with purified protein alone and analysed by fluorescent microscopy as described. The control was done both with a mixture of pseudomurein-containing methanogens and the mixture of bacteria and eukaryotic microorganisms lacking this type of cell wall. Furthermore, controls were included that contained the pseudomurein binding domain-GFP fusion protein to be tested and a mixture of eubacterial and eukaryotic microorganisms.

To determine the ligand of the PeiW CBD, M. thermautotrophicus was harvested by centrifugation and resuspended in 5% (w/v) SDS and boiled for 10 min. Cells were centrifuged and washed twice in M. thermautotrophicus medium. Finally, a binding assay was performed as described above and analysed by fluorescence microscopy.

Microscopic investigations were performed with a Zeiss Axioplan Imaging 2 epifluorescence microscope and an AxioCam HR camera.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

The authors wish to acknowledge Dr A. Wasserfallen (ETH, Zürich, Switzerland) and Dr K. Sandman (Department of Microbiology, Ohio State University, USA) for kindly providing the pET28a expression vectors encoding the N-terminally His-tagged pseudomurein endoisopeptidase PeiW and PeiP. The authors are also thankful for the kind gift of the vector encoding the GFP by Dr H. Alwan (Department Cell Biology, Radboud University Nijmegen, the Netherlands). The authors are also indebted to Boran Kartal for his kind advice with the fluorescence microscope. The work by Nilgün Ayman-Oz was supported by a grant from TUBITAK-BAYG (NATO-A2).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Bateman, A., Coin, L., Durbin, R., Finn, R.D., Hollich, V., Griffiths-Jones, S., et al. (2004) The Pfam protein families database. Nucleic Acids Res 32: D138D141.
  • Bendtsen, J.D., Nielsen, H., Von Heijne, G., and Brunak, S. (2004) Improved prediction of signal peptides: SignalP 3.0. J Mol Biol 340: 783795.
  • Corpet, F. (1988) Multiple sequence alignment with hierarchical clustering. Nucleic Acids Res 16: 1088110890.
  • Corpet, F., Servant, F., Gouzy, J., and Kahn, D. (2000) ProDom and ProDom-CG: tools for protein domain analysis and whole genome comparisons. Nucleic Acids Res 28: 267269.
  • Farkasovska, J., Godany, A., and Vlcek, C. (2003) Identification and characterization of an endolysin encoded by the Streptomyces aureofaciens phage mu 1/6. Folia Microbiol (Praha) 48: 737744.
  • Fricke, W.F., Seedorf, H., Henne, A., Kruer, M., Liesegang, H., Hedderich, R., et al. (2006) The genome sequence of Methanosphaera stadtmanae reveals why this human intestinal archaeon is restricted to methanol and H2 for methane formation and ATP synthesis. J Bacteriol 188: 642658.
  • Hartmann, E., and König, H. (1990a) Comparison of the biosynthesis of the Methanobacterial pseudomurein and the eubacterial murein. Naturwissenschaften 77: 472475.
  • Hartmann, E., and König, H. (1990b) Isolation of lipid activated pseudomurein precursors from Methanobacterium thermoautotrophicum. Arch Microbiol 153: 444447.
  • Hartmann, E., and König, H. (1994) A novel pathway of peptide biosynthesis found in methanogenic archaea. Arch Microbiol 162: 430432.
  • Hartmann, E., König, H., Kandler, O., and Hammes, W. (1990) Isolation of nucleotide activated amino-acid and peptide precursors of the pseudomurein of Methanobacterium thermoautotrophicum. FEMS Microbiol Lett 69: 271275.
  • Kandler, O. (1994) Cell wall biochemistry and 3-domain concept of life. Syst Appl Microbiol 16: 501509.
  • Kandler, O., and König, H. (1978) Chemical composition of peptidoglycan-free cell walls of methanogenic bacteria. Arch Microbiol 118: 141152.
  • Kandler, O., and König, H. (1998) Cell wall polymers in Archaea (Archaebacteria). Cell Mol Life Sci 54: 305308.
  • Kiener, A., König, H., Winter, J., and Leisinger, T. (1987) Purification and use of Methanobacterium wolfei pseudomurein endopeptidase for lysis of Methanobacterium thermoautotrophicum. J Bacteriol 169: 10101016.
  • König, H. (1986) Chemical composition of cell envelopes of methanogenic bacteria isolated from human and animal feces. Syst Appl Microbiol 8: 159162.
  • König, H., and Kandler, O. (1979a) N-acetyltalosaminuronic acid a constituent of the pseudomurein of the genus Methanobacterium. Arch Microbiol 123: 295299.
  • König, H., and Kandler, O. (1979b) Amino acid sequence of the peptide moiety of the pseudomurein from Methanobacterium thermoautotrophicum. Arch Microbiol 121: 271275.
  • König, H., Kralik, R., and Kandler, O. (1982) Structure and modifications of pseudomurein in Methanobacteriales. Zentralbl Bakteriol Mikrobiol Hyg [C] 3: 179191.
  • König, H., Kandler, O., Jensen, M., and Rietschel, E.T. (1983) The primary structure of the glycan moiety of pseudomurein from Methanobacterium thermoautotrophicum. Hoppe Seylers Z Physiol Chem 364: 627636.
  • König, H., Kandler, O., and Hammes, W. (1989) Biosynthesis of pseudomurein – isolation of putative precursors from Methanobacterium thermoautotrophicum. Can J Microbiol 35: 176181.
  • König, H., Hartmann, E., and Karcher, U. (1994) Pathways and principles of the biosynthesis of Methanobacterial cell wall polymers. Syst Appl Microbiol 16: 510517.
  • Kurr, M., Huber, R., König, H., Jannasch, H.W., Fricke, H., Trincone, A., et al. (1991) Methanopyrus kandleri, Gen. and Sp-nov. represents a novel group of hyperthermophilic methanogens, growing at 110 degrees C. Arch Microbiol 156: 239247.
  • Loessner, M.J., Kramer, K., Ebel, F., and Scherer, S. (2002) C-terminal domains of Listeria monocytogenes bacteriophage murein hydrolases determine specific recognition and high affinity-binding to bacterial cell wall carbohydrates. Mol Microbiol 44: 335349.
  • Luo, Y.N., Pfister, P., Leisinger, T., and Wasserfallen, A. (2001) The genome of archaeal prophage Psi M100 encodes the lytic enzyme responsible for autolysis of Methanothermobacter wolfeii. J Bacteriol 183: 57885792.
  • Luo, Y.N., Pfister, P., Leisinger, T., and Wasserfallen, A. (2002) Pseudomurein endoisopeptidases PeiW and PeiP, two moderately related members of a novel family of proteases produced in Methanothermobacter strains. FEMS Microbiol Lett 208: 4751.
  • McGinnis, S., and Madden, T.L. (2004) BLAST: at the core of a powerful and diverse set of sequence analysis tools. Nucleic Acids Res 32: W20W25.
  • Majernik, A.I., Lundgren, M., McDermott, P., Bernander, R., and Chong, J.P.J. (2005) DNA content and nucleoid distribution in Methanothermobacter thermautotrophicus. J Bacteriol 187: 18561858.
  • Makarova, K.S., Aravind, L., and Koonin, E.V. (1999) A superfamily of archaeal, bacterial, and eukaryotic proteins homologous to animal transglutaminases. Protein Sci 8: 17141719.
  • Pennings, J.L.A., Vermeij, P., De Poorter, L.M.I., Keltjens, J.T., and Vogels, G.D. (2000) Adaptation of methane formation and enzyme contents during growth of Methanobacterium thermoautotrophicum (strain Delta H) in a fed-batch fermentor. Antonie Van Leeuwenhoek 77: 281291.
  • Pfister, P., Wasserfallen, A., Stettler, R., and Leisinger, T. (1998) Molecular analysis of Methanobacterium phage Psi M2. Mol Microbiol 30: 233244.
  • Riley, M.L., Schmidt, T., Wagner, C., Mewes, H.W., and Frishman, D. (2005) The PEDANT genome database in 2005. Nucleic Acids Res 33: D308D310.
  • Sambrook, J., Maniatis, T., and Fritsch, E.F. (1989) Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring. Harbor Laboratory Press.
  • Scheffers, D.J., and Pinho, M.G. (2005) Bacterial cell wall synthesis: new insights from localization studies. Microbiol Mol Biol Rev 69: 585607.
  • Slesarev, A.I., Mezhevaya, K.V., Makarova, K.S., Polushin, N.N., Shcherbinina, O.V., and Shakhova, V.V. (2002) The complete genome of hyperthermophile Methanopyrus kandleri AV19 and monophyly of archaeal methanogens. Proc Natl Acad Sci USA 99: 46444649.
  • Smith, D.R., DoucetteStamm, L.A., Deloughery, C., Lee, H.M., Dubois, J., Aldredge, T., et al. (1997) Complete genome sequence of Methanobacterium thermoautotrophicum Delta H: functional analysis and comparative genomics. J Bacteriol 179: 71357155.
  • Stax, D., Hermann, R., Falchetto, R., and Leisinger, T. (1992) The lytic enzyme in bacteriophage Psi-M1-induced lysates of Methanobacterium thermoautotrophicum Marburg. FEMS Microbiol Lett 100: 433438.
  • Steen, A., Buist, G., Leenhouts, K.J., El Khattabi, M., Grijpstra, F., Zomer, A.L., et al. (2003) Cell wall attachment of a widely distributed peptidoglycan binding domain is hindered by cell wall constituents. J Biol Chem 278: 2387423881.
  • Tatusova, T.A., and Madden, T.L. (1999) BLAST 2 SEQUENCES, a new tool for comparing protein and nucleotide sequences. FEMS Microbiol Lett 174: 247250.
  • Wasserfallen, A., Nolling, J., Pfister, P., Reeve, J., and De Macario, E.C. (2000) Phylogenetic analysis of 18 thermophilic Methanobacterium isolates supports the proposals to create a new genus, Methanothermobacter gen. nov., and to reclassify several isolates in three species, Methanothermobacter thermautotrophicus comb. nov., Methanothermobacter wolfeii comb. nov., and Methanothermobacter marburgensis sp nov. Int J Syst Evol Microbiol 50: 4353.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Fig. S1. Titration of M. thermautotrophicus cells with PeiW. Fig. S2. Binding kinetics of PeiW with M. thermautotrophicus cells. Fig. S3. Binding specificity of PeiW. PeiW was mixed with cells of (1) M. thermautotrophicus, (2) E. coli and (3) L. lactis. Table S1. BLAST 2 sequences results for the members of the PB007923 sequence family. Table S2. Descriptive overview of the staining of different pseudomurein-containing archaea with a collection of GFP pseudomure in binding domains. Table S3. Sequences of the primers used in this study (5′ to 3′).

FilenameFormatSizeDescription
mmi5483Figs1Tables1.pdf129KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.