AglP is a S-adenosyl-L-methionine-dependent methyltransferase that participates in the N-glycosylation pathway of Haloferax volcanii

Authors


*E-mail jeichler@bgu.ac.il; Tel. (+972) 8646 1343; Fax (+972) 8647 9175.

Summary

While pathways for N-glycosylation in Eukarya and Bacteria have been solved, considerably less is known of this post-translational modification in Archaea. In the halophilic archaeon Haloferax volcanii, proteins encoded by the agl genes are involved in the assembly and attachment of a pentasaccharide to select asparagine residues of the S-layer glycoprotein. AglP, originally identified based on the proximity of its encoding gene to other agl genes whose products were shown to participate in N-glycosylation, was proposed, based on sequence homology, to serve as a methyltransferase. In the present report, gene deletion and mass spectrometry were employed to reveal that AglP is responsible for adding a 14 Da moiety to a hexuronic acid found at position four of the pentasaccharide decorating the Hfx. volcanii S-layer glycoprotein. Subsequent purification of a tagged version of AglP and development of an in vitro assay to test the function of the protein confirmed that AglP is a S-adenosyl-L-methionine-dependent methyltransferase.

Introduction

Of the various post-translational modifications proteins can experience, N-glycosylation is among the most prominent. However, whereas the eukaryal and bacterial N-glycosylation pathways are well defined (for reviews, see Helenius and Aebi, 2004; Szymanski and Wren, 2005; Weerapana and Imperiali, 2006), comparatively less is known of this protein-processing event in Archaea (Eichler and Adams, 2005). Recent studies on N-glycosylation in the halophile Haloferax volcanii, the methanogens Methanococcus voltae and Methanococcus maripaludis and the thermophile Pyrococcus furiosus have begun to correct this situation (Abu-Qarn et al., 2008a; Igura et al., 2008; Yurist-Doutsch et al., 2008a; VanDyke et al., 2009).

In Hfx. volcanii, products of the agl (archaeal glycosylation) genes have been shown to mediate the N-glycosylation of the surface (S)-layer glycoprotein, a reporter of this post-translational modification (Sumper et al., 1990). In the S-layer glycoprotein, at least two asparagine residues are modified by a pentasaccharide comprising two hexoses, two hexuronic acids and a 190 Da saccharide, corresponding to either a dimethylated hexose or a methyl ester of hexuronic acid (Abu-Qarn et al., 2007). To date, Agl proteins involved in the assembly (i.e. AglD, AglE, AglF, AglG, AglI, AglJ and AglM) and attachment (i.e. AglB) of this oligosaccharide have been identified (Abu-Qarn and Eichler, 2006; Abu-Qarn et al., 2007; 2008b; Yurist-Doutsch et al., 2008b; 2010). While the role of AglB as an oligosaccharyltransferase and the roles of AglD, AglE, AglG, AglI and AglJ as glycosyltransferases have been inferred from gene deletion-based studies (Abu-Qarn and Eichler, 2006; Abu-Qarn et al., 2007; 2008b; Yurist-Doutsch et al., 2008b), AglF and AglM have been directly shown to act as a UDP-glucose pyrophosphorylase and a UDP-glucose dehydrogenase, respectively (Yurist-Doutsch et al., 2010).

With the exception of aglD, involved in adding the final hexose to the pentasaccharide decorating the S-layer glycoprotein (Abu-Qarn and Eichler, 2006), the genes encoding the various agl genes are clustered within an island in the Hfx. volcanii genome stretching from HVO_1517 (aglJ) to HVO_1531 (aglM) (Yurist-Doutsch and Eichler, 2009; Yurist-Doutsch et al., 2010). The agl cluster includes genes initially identified through their similarities to known components of the eukaryal or bacterial N-glycosylation pathways, genes neighbouring these sequences as well as novel genes discovered upon manual reannotation of this region of the genome (Abu-Qarn and Eichler, 2006; Yurist-Doutsch and Eichler, 2009). For example, the product of HVO_1522 was annotated as AglP, based on its proximity to genes encoding experimentally verified components of the Hfx. volcanii N-glycosylation pathway (Yurist-Doutsch and Eichler, 2009). Relying on homology-based analysis, the same study predicted AglP to act as a methyltransferase.

In the present report, AglP was purified and shown to function as a S-adenosyl-L-methionine (SAM)-dependent methyltransferase. Furthermore, AglP was shown to be responsible for the methylation that yields the 190 Da saccharide found at position four of the N-linked pentasaccharide decorating the Hfx. volcanii S-layer glycoprotein, now revealed to be a methyl ester of hexuronic acid.

Results

Deletion of Hfx. volcanii aglP modulates N-glycosylation of the S-layer glycoprotein

To address the participation of aglP in Hfx. volcanii protein glycosylation, the encoding gene was deleted according to the so-called ‘pop-in/pop-out’ approach developed by Allers et al. (2004). In this approach, the aglP sequence within the genome of cells of the Hfx. volcanii WR536 (H53) parent strain, a uracil and tryptophan auxotroph, is replaced by the tryptophan synthase-encoding Hfx. volcanii trpA gene (HVO_0789) via homologous recombination and appropriate selection through the use of the pyrE-containing plasmid, pTA131, and plating onto casamino acids lacking uracil and tryptophan. PCR amplification allows one to follow the genomic integration of the trpA-introducing plasmid, as well as the subsequent expulsion of the plasmid, along with aglP that replaced the trpA sequence originally contained in the plasmid. Use of a primer pair directed against internal and downstream flanking regions of aglP yielded a PCR amplification product in the parent strain but not in the deletion strain (Fig. 1A, primer pair a, compare left and middle panels). By contrast, a primer pair directed against the start of trpA and the flanking region downstream to aglP only yielded a PCR amplification product in the deletion strain (Fig. 1A, primer pair b, compare left and middle panels). Deletion of the gene was further confirmed when PCR amplification was performed using genomic DNA from the parent and deletion strains as template, together with primers directed against the aglP coding region (Fig. 1A, primer pair c, right panel).

Figure 1.

aglP is not essential for Hfx. volcanii survival.
A. Left and middle panels: PCR amplification was performed using a forward primer directed to a sequence within the aglP coding region and a reverse primer directed at the aglP 3′ flanking region (primer pair a) or using a forward primer directed to a sequence at the start of the trpA sequence and the same reverse primer as above (primer pair b), together with genomic DNA from cells of the parent strain (parent; left panel) or from cells where aglP had been replaced with trpAaglP; middle panel), as template. Right panel: PCR amplification was performed using primers pair c directed against the aglP coding region, together with genomic DNA from cells of the parent strain (parent) or the aglP-deleted strain (ΔaglP) as template. The positions to which the various primer pairs bind are shown in the drawing below the panels. Note that aglP and trpA are surrounded by the same flanking regions.
B. RT-PCR was performed using primers directed at aglP (top row of panels) or aglI (bottom row of panels) together with cDNA (left lane of each panel) or RNA (right lane of each panel) from the parent strain (parent; left column of panels) or aglP-deleted cells (ΔaglP; right columns of panels) as template. The identity of the PCR products was confirmed by sequencing. The positions to which the various primer pairs bind, as well as the relative positions of aglP, aglI and intervening agl genes, are shown in the drawing below the panels. In both drawings, all gene sequences are drawn as being of equivalent length for simplicity.

The absence of aglP in the deletion strain was also verified at the RNA level by reverse-transcription (RT)-PCR, performed as described previously (Abu-Qarn and Eichler, 2006). Here, RNA or cDNA from the parent or deletion strains served as template for PCR amplification, together with primers directed against the coding region of aglP or aglI, a known component of the Hfx. volcanii N-glycosylation pathway (Yurist-Doutsch et al., 2008b), serving as a positive control. As reflected in Fig. 1B, no PCR products were obtained when cDNA from the deletion strain served as template in a reaction involving primers directed against aglP. By contrast, PCR products were readily obtained when the same reactions were repeated using cDNA from the parent or deletion strain as template together with primers directed against aglI (or aglQ or aglE; not shown).

Having confirmed the deletion of aglP at the DNA and RNA levels, efforts next focused on determining whether the absence of AglP affected N-glycosylation in Hfx. volcanii. Accordingly, mass spectrometry (MS) was employed to compare the N-linked glycan profile of the S-layer glycoprotein, a reporter of this post-translational modification, in cells of the parent strain and in cells lacking AglP (Fig. 2). In agreement with earlier results (Abu-Qarn et al., 2007), the MS spectrum of a S-layer glycoprotein-derived peptide containing Asn-13 showed it to be decorated with a pentasaccharide, together with its biosynthetic precursors, comprising a hexose (glycopeptide at m/z 1743), two hexuronic acids (glycopeptides at m/z 1919 and 2095, respectively), a 190 Da species (glycopeptide at m/z 2285) and a final hexose subunit (glycopeptide at m/z 2448). In the case of the same peptide derived from the S-layer glycoprotein of the aglP deletion strain, a smaller glycan was observed. Here, a tetrasaccharide was the largest glycan, with the glycopeptides being found to contain the Asn-linked hexose (m/z 1743), the two hexuronic acids (m/z 1919 and 2095, respectively) and finally, a 176 Da species (m/z 2271), rather than the 190 Da observed in the parent strain. Thus, in addition to lacking the final hexose subunit, the S-layer glycoprotein-attached glycan moiety from cells of the deletion strain presents a subunit at position four of the pentasaccharide that is 14 Da lighter than the 190 Da entity normally found at this position.

Figure 2.

MALDI-TOF MS analysis of the Asn-13-containing Hfx. volcanii S-layer glycoprotein-derived tryptic glycopeptide. The MALDI-TOF spectra of the Asn-13-containing tryptic peptide derived from the S-layer glycoprotein from the parent (left panel; parent) and aglP-deleted strains (right panel; ΔaglP) are shown. The components of the glycopeptide-associated sugar residues are shown in the inset box, while the glycan moieties decorating the peptide peaks are marked on the spectra, accordingly. In the ΔaglP panel, the arrow shows the position of the peptide modified by the tetrasaccharide that includes a novel subunit at position four.

The 190 Da species found at position four of the N-linked S-layer glycoprotein pentasaccharide is a methyl ester of hexuronic acid

Previous analysis had assigned the 190 Da species at position four of the S-layer glycoprotein-linked pentasaccharide to likely correspond to either a dimethylated hexose or to a methyl ester of hexuronic acid (Abu-Qarn et al., 2007). As such, it would appear that in cells lacking AglP, the 176 Da saccharide found at position four of the pentasaccharide is lacking a methyl group, as observed by the decrease in mass of this reporter glycopeptides by 14 mass units, yielding either methylated hexose or hexuronic acid. To distinguish between these possibilities, methyl esterification of the Asn13-containing S-layer glycoprotein-derived peptide from the aglP-deleted cells was performed, as described previously (Abu-Qarn et al., 2007). As a result of such treatment, carboxylic acid groups are converted to their methyl esters, reflected as a 14 Da mass shift for each carboxylic acid moiety. As previously demonstrated, the trisaccharide-bearing Asn-13-containing S-layer glycoprotein peptide derived from the parent strain (Abu-Qarn et al., 2007), showed a shift of m/z 98 following esterification, explained by the transformation of the seven carboxylic acid groups found on the two Glu, the two Asp and the C-terminal residues of the peptide, as well as on the two hexuronic acids found at positions two and three of the N-linked saccharide (Table 1). Given that no further shift was observed upon similar treatment of the tetrasaccharide-bearing peptide derived from the same source, the 190 Da sugar species found at position four of the N-linked pentasaccharide was deemed as not presenting a free carboxyl group.

Table 1.  Shifts in glycopeptide position following methyl esterification.
Asn-13-containing peptide modified byNative m/zm/z following methyl esterificationShift in m/z in multiples of 14
  1. n/a, not applicable.

Parent strain   
 Monosaccharide174318135
 Disaccharide191920036
 Trisaccharide209521937
 Tetrasaccharide228523837
 Pentasaccharide244725457
ΔaglP strain   
 Monosaccharide174318135
 Disaccharide191920036
 Trisaccharide209521937
 Tetrasaccharide227123838
 Pentasacchariden/an/an/a

When methyl-esterification of the Asn-13-containing S-layer glycoprotein peptide derived from aglP-deleted cells was performed (Fig. 3A, Table 1), the mono-, di- and trisaccharide-bearing peptides were shifted 70, 84 and 98 Da respectively, consistent with the same glycopeptides derived from the parent strain, reflecting the presence of five, six and seven target carboxylic groups, respectively (Abu-Qarn et al., 2007). However, unlike the tetrasaccharide-bearing glycopeptide from the parent strain, the m/z 2271 peak from the AglP-lacking cells was shifted by 112 mass units to m/z 2383 position, reflecting the presence of eight carboxylic moieties. This corresponds to an additional methyl group on the tetrasaccharide glycan derived from the AglP-lacking cells. It can thus be concluded that the 176 Da species found at position four of the tetrasaccharide linked to the S-layer glycoprotein derived from the aglP-deleted cells corresponds to a hexuronic acid. By extension, the 190 Da saccharide found at position four of the pentasaccharide N-linked to the S-layer glycoprotein in the parent strain corresponds, therefore, to the methyl ester of a hexuronic acid (Fig. 3B). The clusters of signals below each of the annotated molecular ions (Fig. 3A) are due to incomplete esterification of a portion of the sample under the conditions used and/or lactonization (intervals of 14 and 18 Da, respectively).

Figure 3.

Methyl esterification, using methanolic HCl, of the Asn-13-containing Hfx. volcanii S-layer glycoprotein-derived tryptic glycopeptide from cells lacking AglP confirms the 190 Da saccharide of the peptide-linked pentasaccharide is a methylated ester of hexuronic acid.
A. The Asn-13-containing Hfx. volcanii S-layer glycoprotein-derived glycopeptide from cells lacking AglP was incubated with 1 M MeOH-HCl, as previously described (Abu-Qarn et al., 2007). The molecular ions of the esterified peptide modified by the mono-, di- and trisaccharides, as well as by the tetrasaccharide terminating in a 176 Da unit, are indicated by arrows. The numbers above each arrow indicate the number of carboxylic acids converted to their methyl esters, with each such conversion adding 14 Da to the mass of the peptide.
B. Schematic depiction of the pentasaccharide attached to the S-layer glycoprotein. N – modified asparagine residue; Hex – hexose; HexUA – hexuronic acid.

AglP is a SAM-dependent methyltransferase that modifies the hexuronic acid found at position four of the S-layer glycoprotein-linked pentasaccharide

Given the MS results pointing to AglP as being responsible for methylation of the sugar subunit found at position four of the pentasaccharide decorating the Hfx. volcanii S-layer glycoprotein, efforts were directed at biochemically confirming that AglP is a methyltransferase. Initial support for AglP being a methyltransferase came from an earlier homology-based prediction (Yurist-Doutsch and Eichler, 2009) in which analysis of the deduced 239-residue AglP sequence at InterProScan ( http://www.ebi.ac.uk/Tools/InterProScan/) revealed the protein to contain a sequence motif between residues 69 and 201 assigning it to the IPR006342 FkbM methyltransferase family, exemplified by FkbM from Streptomyces strain MA6548 (Motamedi et al., 1996). Now, to experimentally verify this prediction, Hfx. volcanii cells were transformed to express AglP fused to the Clostridium thermocellum cellulose-binding domain (CBD). This CBD serves as a purification tag capable of interacting with cellulose even in hypersaline conditions, such as those in which Hfx. volcanii grow (Irihimovitch et al., 2003). As reflected in Fig. 4A, the cellulose-based purification technique (see Experimental procedures) led to the isolation of a single protein species, migrating close to the predicted molecular mass of the CBD-AglP chimera (46 346 Da).

Figure 4.

Purification and sub-cellular localization of CBD-AglP.
A. CBD-AglP can be purified from Hfx. volcanii cells transformed to express the chimera. A Coomassie-stained SDS-PAGE gel showing molecular weight markers (lane 1), an aliquot of a total Hfx. volcanii protein extract (lane 2) and cellulose-purified CBD-AglP (lane 3) is presented.
B. CBD-AglP is localized to the cytosol. Hfx. volcanii cells expressing CBD-AglP were broken by sonication, membrane and cytoplasmic fractions were obtained by ultracentrifugation and the different fractions were Coomassie-stained to reveal the presence of the S-layer glycoprotein (SLG, migrating as a 180–200 kDa protein (Sumper et al., 1990); top panel) or immunoblotted with antibodies against SRP54 (migrating as a 50 kDa protein (Tozik et al., 2002); middle panel) or CBD (lower panel).

Next, the sub-cellular localization of AglP was considered. Bioinformatics-based analysis of the AglP sequence using the PRED-SIGNAL ( http://bioinformatics.biol.uoa.gr/PRED-SIGNAL/; Bagos et al., 2009) and TMHMM (http://www.cbs.dtu.dk/services/TMHMM-2.0/) assign a cytoplasmic localization to AglP, with these servers failing to, respectively, detect either the presence of an archaeal signal peptide or any trans-membrane domains. The protein was similarly assigned to the Hfx. volcanii cytoplasm by other sub-cellular localization prediction algorithms found at the ExPASy Proteomics Server ( http://www.expasy.ch). To experimentally verify these predictions, Hfx. volcanii cells transformed to express CBD-AglP were subjected to sub-cellular fractionation and the distribution of the chimera was assessed by immunoblot using antibodies raised against the CBD moiety. To confirm the efficacy of fractionation, the soluble and pellet fractions were probed for the presence of SRP54 (Tozik et al., 2002) and the S-layer glycoprotein, markers of the cytoplasm and membrane, respectively. Like SRP54, CBD-AglP was essentially restricted to the soluble phase of the cell (Fig. 4B). By contrast, the S-layer glycoprotein was largely confined to the membrane phase.

AglP methyltransferase activity was determined by assaying the ability of the purified cellulose-bound, CBD-tagged protein to transfer a [3H]-methyl group from [3H]-methyl-SAM to membrane fragments prepared from Hfx. volcaniiΔaglP cells. As reflected in Fig. 5A, after subtracting the background level of non-specifically-bound radioactivity associated with untreated cellulose beads (∼3800 c.p.m.), almost fourfold more radioactivity could be counted in membrane fragments incubated in the presence of [3H]-methyl-SAM and cellulose-bound CBD-AglP than in the same reaction mixture instead containing cellulose-bound CBD-AglD (Plavner and Eichler, 2008), serving as a negative control. The radioactivity associated with the membrane fragments prepared from Hfx. volcaniiΔaglP cells following incubation with cellulose-bound CBD-AglD above that value associated with non-specific binding to the cellulose (i.e. ∼8500  c.p.m.) is also likely due to non-specific binding of [3H]-methyl-SAM.

Figure 5.

AglP is a SAM-dependent methyltransferase.
A. The ability of cellulose-bound CBD-AglP to transfer a [3H]-methyl group from [3H]-methyl-SAM to membrane fragments prepared from Hfx. volcaniiΔaglP cells was tested as described in the Experimental procedures section. In control experiments, cellulose-bound CBD-AglD or untreated cellulose beads were employed. A representative experiment of four is shown, with each bar corresponding to the mean value of triplicate samples ± standard deviation.
B. The experiment described in A was repeated with cellulose-bound CBD-AglP and increasing amounts (0–0.5 mM) of unlabelled SAM. A representative experiment of three is shown. In each column, the background reading obtained in the same experiment upon incubation of membrane fragments prepared from Hfx. volcaniiΔaglP cells with untreated cellulose beads was subtracted.
C. The ability of cellulose-bound CBD-AglP to perform [3H]-methylation of membrane fragments prepared from Hfx. volcaniiΔaglP or ΔaglE cells upon [3H]-methyl-SAM addition was tested. In control experiments, untreated cellulose beads were combined with membrane fragments prepared from Hfx. volcaniiΔaglP cells and [3H]-methyl-SAM. Each bar represents the average of three samples ± standard deviation.

To further confirm the specific nature of the AglP methyltransferase activity observed, membrane fragments prepared from Hfx. volcaniiΔaglP cells were again incubated with cellulose-bound CBD-AglP and [3H]-methyl-SAM, only this time in the presence of increasing amounts of unlabelled SAM (0–0.5 mM). As the amount of unlabelled SAM in the reaction mixture was increased, less radioactive label was incorporated into the membrane fraction (Fig. 5B). Indeed, if the extent of [3H]-methylation of membrane fragments prepared from Hfx. volcaniiΔaglP cells challenged with [3H]-methyl-SAM alone is taken as 100%, it was determined that only 22% of that amount of radiolabel was introduced when 0.5 mM SAM was also included in the reaction.

Next, experiments were undertaken to confirm that the methyl group added by AglP indeed modified the hexuronic acid found at position four of the oligosaccharide that eventually decorates selected Asn residues of the S-layer glycoprotein to yield the 190 Da methyl ester of hexuronic acid previously identified at this position (Abu-Qarn et al., 2007), thus concurring with the MS results presented above (see Figs 2 and 3). Accordingly, the ability of CBD-AglP to catalyze the transfer of a [3H]-methyl group from [3H]-methyl-SAM to membrane fragments of Hfx. volcanii cells deleted of aglE was considered. AglE has been previously shown to participate in the addition of the sugar subunit found at position four of the pentasaccharide decorating the S-layer glycoprotein (Abu-Qarn et al., 2008b). Hence, if AglP truly modifies the hexuronic acid found at pentasaccharide position four, then no transfer of a [3H]-methyl group should be observed in Hfx. volcanii membrane fragments prepared from cells lacking AglE. Figure 5C shows the level of radiolabelling of membrane fragments prepared from Hfx. volcaniiΔaglE cells was comparable to the background level of radiolabelling attained when the reaction contained untreated cellulose beads, confirming that methylation of the S-layer glycoprotein-linked pentasaccharide occurs on subunit four in an AglP-mediated manner.

Discussion

With the involvement of eight Agl proteins (i.e. AglB, AglD, AglE, AglF, AglG, AglI, AglJ and AglM) in the N-glycosylation of the Hfx. volcanii S-layer glycoprotein having already been demonstrated (Abu-Qarn and Eichler, 2006; Abu-Qarn et al., 2007; 2008b; Yurist-Doutsch et al., 2008b; 2010) and other Agl proteins having been identified (Yurist-Doutsch and Eichler, 2009), efforts have begun to focus on defining the specific functions of individual Agl proteins. Accordingly, AglF was shown to be a UDP-glucose pyrophosphorylase, while AglM was verified to serve as a UDP-glucose dehydrogenase (Yurist-Doutsch et al., 2010). In a coupled in vitro assay, the ability of the two proteins to cooperate in the biogenesis of hexuronic acids (preferentially glucuronic acid), such as found at position three of the pentasaccharide decorating the Hfx. volcanii S-layer glycoprotein, was revealed (AglM apparently cooperates with a different UDP-hexose pyrophosphorylase to generate the hexuronic acid found at pentasaccharide position two; Abu-Qarn et al., 2007; Yurist-Doutsch et al., 2010). In the present report, experiments conducted at both the gene and protein levels demonstrated that AglP is a SAM-dependent methyltransferase responsible for methylating the hexuronic acid found at position four of the N-linked pentasaccharide decorating the Hfx. volcanii S-layer glycoprotein.

Deletion of the AglP-encoding gene resulted in a S-layer glycoprotein modified by a tetrasaccharide rather that the pentasaccharide detected on the native protein. Moreover, the tetrasaccharide observed by MS in the ΔalgP cells contained a 176 Da species at position four, rather than the 190 Da species found at this site in the native protein. This difference of 14 Da likely reflects the contribution of a methyl group, because previous efforts (Abu-Qarn et al., 2007) had identified the 190 Da species as corresponding to either a methyl ester of hexuronic acid or dimethylated hexose. Methanolic HCl-based methyl esterification prior to MS analysis of the carboxylic groups present in an Asn-13-containing S-layer glycoprotein-derived peptide from cells lacking AglP confirmed that position four of the N-linked oligosaccharide corresponds to a methyl ester of hexuronic acid. Such findings indicate that AglP functions as a methyltransferase, as originally suggested based on sequence homology considerations (Yurist-Doutsch and Eichler, 2009). Relying on an in vitro assay for AglP developed in this study, the SAM-dependent methyltransferase activity of the protein was directly shown.

It, moreover, appears that AglP-mediated methylation occurs on the lipid-linked tetrasaccharide precursor of the pentasaccharide eventually transferred to the S-layer glycoprotein, because any AglP-methylated soluble hexuronic acid would be expected to also be incorporated at pentasaccharide positions two, three or four, a situation which is not observed. Given the apparent cytoplasmic localization of AglP, it is also unlikely that the methyltransferase modifies the oligosaccharide after its transfer to the S-layer glycoprotein, a process thought to occur on the external surface of the cell (Lechner and Wieland, 1989). Indeed, the active site of AglD, responsible for adding the final hexose to the pentasaccharide transferred to the S-layer glycoprotein, faces the cytoplasm (Plavner and Eichler, 2008).

The methylation of the fourth saccharide subunit of the pentasaccharide that is ultimately N-linked to the Hfx. volcanii S-layer glycoprotein, as reported here, is not the first example of such modification of an oligosaccharide eventually delivered to an archaeal glycoprotein. In Halobacterium salinarum (then called Halobacterium halobium), SAM-dependent methylation of a glucose subunit found as part of a lipid-linked sulphated oligosaccharide intermediate involved in N-glycosylation was reported (Lechner et al., 1985). It was further shown that inhibition of such methylation prevented transfer of the oligosaccharide to its target protein, likely the S-layer glycoprotein and two lower molecular weight proteins in this species. Although the same sulphated oligosaccharide as linked to the dolichylphosphate carrier is also found on the Hbt. salinarum S-layer glycoprotein, methylation of the glucose unit in question is only detected in the lipid-linked oligosaccharide and hence, appears to represent a transient modification. By contrast, a mannose subunit of the hexasaccharide shown to decorate select Asn residues of the Methanothermus fervidus S-layer glycoprotein is methylated (Kärcher et al., 1993). In neither case is the role of oligosaccharide methylation clear, although in the former, it has been proposed that such modification is important for translocation of the lipid-linked oligosaccharide across the plasma membrane (Lechner et al., 1985). In Hfx. volcanii, it was shown in the present study that the S-layer glycoprotein is modified by a tetrasaccharide in cells deleted of aglP, and hence incapable of oligosaccharide methylation, arguing against a role for methylation in delivering the lipid-linked oligosaccharide across the membrane in the N-glycosylation pathway of this archaeon.

Although methylated saccharides are detected in the N-linked glycans of eukaryal glycoproteins, methylated intermediates have not been reported in the eukaryal N-glycosylation pathway. Similarly, the only bacterial N-glycosylation pathway defined to date, namely that of Campylobacter jejuni, does not implicate a methyltransferase (Linton et al., 2005; Szymanski and Wren, 2005; Weerapana and Imperiali, 2006). Nonetheless, a blast search detects homologues of AglP not only in other Archaea but also in numerous bacterial species, including Campylobacter fetus ssp. fetus 82–40. However, it remains to be determined whether these species are capable of performing N-glycosylation, and if so, whether the process involves a methylation step. In the same blast search, no eukaryal homologues of Hfx. volcanii AglP were detected. Indeed, only after five iterations of psi-blast was a single eukaryal AglP homologue detected in the dinoflagellate, Karlodinium micrum.

The pentasaccharide decorating at least two Asn residues of the Hfx. volcanii S-layer glycoprotein is known to contain a linking hexose residue, two hexuronic acids [apparently glucuronic acids (Yurist-Doutsch et al., 2010)], a methyl ester of hexuronic acid (also likely to be glucuronic acid) and a final hexose. Work aimed at delineating the precise identities of these sugars is in progress, as are studies seeking to determine which other potential N-glycosylation sites of the protein are modified, and if so, whether by the same pentasaccharide.

Experimental procedures

Strains and growth conditions

The Hfx. volcanii parent strain WR536 (H53) and the same strain deleted of aglP were grown in complete medium containing 3.4 M NaCl, 0.15 M MgSO47H2O, 1 mM MnCl2, 4 mM KCl, 3 mM CaCl2, 0.3 % (w/v) yeast extract, 0.5 % (w/v) tryptone, 50 mM Tris-HCl, pH 7.2, at 40°C (Mevarech and Werczberger, 1985). Hfx. volcanii cells deleted of aglE were described previously (Abu-Qarn et al., 2008b).

Deletion of aglP

Deletion of Hfx. volcanii aglP (accession number CAW30727) was achieved as previously described (Allers et al., 2004; Abu-Qarn and Eichler, 2006). To amplify regions of approximately 500 bp in length flanking the coding sequence of aglP, the AglP5′upfor (GGGctcgagGCAAACGACGTCTGAGGAAC) and AglP5′uprev (CCCaagcttATATTTAAGATGAAGACTGA) primers, directed against the upstream flanking region, and the AglP3′downfor (GGGggatccATACACTATCTAGTTAGTGAC) and AglP3′downrev (CCCtctagaATCGGGCGCGCTGGGAGGTC) primers, directed against the downstream flanking region, were employed. XhoI and HindIII sites were introduced in the AglP5′up and AglP5′uprev sequences respectively, while BamHI and XbaI sites were introduced in the AglP3′down and AglP3′downrev sequences respectively (introduced sites are listed in each primer in lower case letters). To confirm deletion of aglP at the DNA level, PCR amplification was performed using primers against an internal region of aglP (AglPfor; ATGACAATAGTTAAAAAAGTGGCG) or the start of trpA[CCCgaattcTTATGTGCGTTCCGGATGCG; including an HindIII (lower case letters) introduced during an earlier cloning step (see Abu-Qarn and Eichler, 2006)], each together with a reverse primer directed against the region downstream of aglP (AglP3′downrev) [yielding primer pairs a and b, respectively (see Fig. 1)], or using primers AglPfor and AglPrev (TTAATCATTTTGTCTGCGACCAATAAC), designed to amplify the aglP coding region [primer pair c (see Fig. 1)]. RT-PCR was performed as described previously (Abu-Qarn and Eichler, 2006), using primers AglPfor and AglPrev to test for aglP transcription or primers AglIfor (GCTGATTCTCCGTTTC) and AglIrev (AGCGGGTGTTCCCGC) to test for aglI transcription.

Mass spectrometry of the S-layer glycoprotein

For in-gel tryptic digestion of the Hfx. volcanii S-layer glycoprotein from cells of the parent strain, or the same strain depleted of aglP, samples were run on 10% pre-cast gels (Invitrogen, Paisley, UK) and stained with Novex Colloidal blue stain (Invitrogen). The bands of interest were excised, destained in 400 µl of 50% (v/v) acetonitrile in 0.1 M ammonium bicarbonate, pH 8.4, and dried using a SpeedVac drying apparatus. The gel slices were rehydrated in 20 µl of trypsin working solution (Promega sequencing grade modified trypsin, prepared according to the manufacturer's instructions) and incubated at 37°C, overnight. The supernatant was removed and digestion was terminated by addition of 50 µl of 0.1% (v/v) trifluoroacetic acid (10 min, 37°C). The supernatant was removed and the peptides were further extracted with 200 µl of 60% (v/v) acetonitrile/0.1% (v/v) trifluoroacetic acid (15 min, 37°C). The supernatant was again removed and pooled with the previous supernatant. Both extraction steps were then repeated and the supernatants pooled. The volume of the combined supernatants was subsequently concentrated using a SpeedVac drying apparatus.

For offline liquid chromatography matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) analysis, tryptic peptides were separated using the Ultimate 3000 LC system (Dionex, Sunnyvale, CA), fitted with a Pepmap analytical C-18 nanocapillary (75 µm internal diameter × 15 cm length; Dionex). The digest was loaded onto the column and eluted using solvent A [0.1% (v/v) trifluoroacetic acid in 2% (v/v) acetonitrile and solvent B (0.1% (v/v) trifluoroacetic acid in 90% (v/v) acetonitrile), using a gradient of 0–60% solvent B (0–36 min). Eluting fractions were mixed with α-cyano-hydroxy cinnamic acid matrix and spotted onto a metal MALDI target plate. MALDI-TOF MS was performed using an Applied Biosystems 4800 mass spectrometer in the positive reflectron mode and set for delayed extraction. MS/MS was performed with the CID setting turned on. Sequazyme peptide mass standards were used as external calibrants. Esterification of carboxylic groups using methanolic HCl was also performed as previously described (Abu-Qarn et al., 2007). Briefly, glycopeptides were incubated with 2 M HCl/MeOH for 15 min, purified by nanoLC, and then analysed by MALDI-TOF.

Construction of a CBD–AglP chimera

To generate a plasmid encoding Clostridium thermocellum CBD-tagged AglP (CBD-AglP), the aglP gene was PCR amplified using primers (forward primer: GGGcatatgACAATAGTTAAAAAAGTGGC; reverse primer: CCCggtaccTTAATCATTTTGTCTGCGAC) designed to introduce NdeI and KpnI restriction sites (in lower case letters in the primer sequences) at the start and end of the gene, respectively. The amplified fragment was digested with NdeI and KpnI and ligated into plasmid pWL-CBD (Irihimovitch et al., 2003), previously digested with the same restriction enzymes, to produce plasmid pWL-CBD-AglP. Plasmid pWL-CBD-AglP was introduced into cells of the Hfx. volcanii WR536 (H53) parent strain.

Sub-cellular fractionation

Haloferax volcanii cells (1 ml) were broken by sonication (2 s on and 1 s off for 90 s, 25% output, Misonix XL2020 ultrasonicator). Unbroken cells were pelleted in a microfuge (9000 g, 10 min, 4°C) and the resulting supernatant was centrifuged in an ultracentrifuge (Sorvall M120; 240 000 g, 12 min, 4 °C). While the resulting supernatant was directly precipitated in 15% (w/v) tri-chloroacetic acid, the pelleted membrane fraction was resuspended in 200 µl of distilled water and then precipitated in 15% (w/v) tri-chloroacetic acid. Proteins were electrotransferred from SDS-PAGE gels to nitrocellulose membranes (0.45 µm, Schleicher & Schuell, Dassel, Germany) and incubated with polyclonal rabbit anti-CBD antibodies (a gift from Ed Bayer, Weizmann Institute of Science; 1:10 000 dilution) or anti-SRP54 antibodies (Tozik et al., 2002; 1:1000). Horseradish peroxidase-conjugated goat anti-rabbit antibodies (Bio-Rad), serving as secondary antibodies, were used at a 1:4000 dilution. Antibody binding was detected using ECL Western blotting detection reagent (Amersham, Bucks, UK). The distribution of the S-layer glycoprotein was determined by Coomassie staining of the cytosolic and membrane protein pools in SDS-PAGE gels.

S-adenosyl-L-methionine-dependent methyltransferase assay

Cellulose-based purification of CBD-AglP from the cytosolic fraction of Hfx. volcanii cells (1 ml) transformed to express the chimera was performed essentially as described previously for CBD-AglD (Plavner and Eichler, 2008). Briefly, transformed cells (1 ml) were centrifuged (3000 g, 3 min, room temperature) and resuspended in 1 ml of lysis buffer [1% Triton X-100 (v/v), 1.8 M NaCl, 50 mM Tris-HCl, pH 7.2] containing 1 mM PMSF. The mixtures were rocked (10 min, room temperature), after which time 50 µl of a 10% (w/v) solution of cellulose beads was added. After a 20 min rocking at room temperature, the suspension was centrifuged (3000 g, 3 min, room temperature), the supernatant was discarded and the cellulose pellet was washed with 2 M NaCl, 50 mM Tris-HCl, pH 7.2. This washing procedure was repeated twice. To test the methyltransferase activity of the purified protein, 50 µl of the cellulose-bound CBD-AglP slurry was added to a microfuge tube containing 198 µl buffer (2 M NaCl, 50 mM Tris-HCl, pH 7.2) and 2 µl [3H]-methyl-SAM (15 Ci mmol–1; Amersham). In some cases, the buffer also contained 0–0.5 mM of unlabelled SAM (Sigma, St Louis, MO). To this mixture, 50 µl of membranes prepared from 1 ml of Hfx. volcanii cells deleted of aglP and resuspended in 2 M NaCl, 50 mM Tris-HCl, pH 7.2, was added. In control experiments, cellulose beads containing bound CBD-AglD (20) or untreated cellulose beads were employed. Following incubation (90 min, 37°C), the cellulose beads were precipitated (5000 r.p.m. in a microfuge, 4°C) and 200 µl of the supernatant, containing the membrane fraction, were removed and captured on GF/C glass fibre filters (25 mm diameter; Whatman, Kent, UK). After extensive washing to remove unprocessed [3H]-methyl-SAM, the dried filters were added to vials containing 4 ml scintillation fluid and bound radioactivity was determined in a β-counter.

Acknowledgements

J.E. is supported by grants from the Israel Science Foundation (Grant 30/07) and the US Air Force Office for Scientific Research (Grant FA9550-07-10057). A.D. was supported by the Biotechnology and Biological Sciences Research Council (Grants BBF008309 and BBC5196701). S.Y.D. is the recipient of a Negev-Faran Associates Scholarship.

Ancillary