Analysis of Agl Proteins: Components of the N-Glycosylation Process In

Analysis of Agl Proteins: Components of the N-glycosylation Process in

Haloferax volcanii

Thesis submitted in partial fulfillment of the requirements for the degree of “DOCTOR OF PHILOSOPHY”

By Lina Kaminski

Submitted to the Senate of Ben-Gurion University of the Negev

November, 2013

Beer-Sheva, Israel

Analysis of Agl Proteins: Components of the N-glycosylation Process in

Haloferax volcanii

Thesis submitted in partial fulfillment of the requirements for the degree of “DOCTOR OF PHILOSOPHY”

By Lina Kaminski

Submitted to the Senate of Ben-Gurion University of the Negev

Approved by the advisor Prof. Jerry Eichler______

Approved by the Dean of the Kreitman School of Advanced Graduate Studies______

November, 2013 Beer-Sheva, Israel

This work was carried out under the supervision of Prof. Jerry Eichler

In the Department of Life Sciences

Faculty of Natural Sciences

Research-Student's Affidavit when Submitting the Doctoral Thesis for Judgment

I, Lina Kaminski, whose signature appears below, hereby declare that (Please mark the appropriate statements):

√ I have written this Thesis by myself, except for the help and guidance offered by my Thesis Advisor.

√ The scientific materials included in this Thesis are products of my own research, culled from the period during which I was a research student.

√ This Thesis incorporates research materials produced in cooperation with others, excluding the technical help commonly received during experimental work. Therefore, I am attaching another affidavit stating the contributions made by myself and the other participants in this research, which has been approved by them and submitted with their approval.

Date: Student's name: Lina Kaminski Signature:______

I would like to thank my advisor, Prof. Jerry Eichler.

Thank you for your guidance and constant support during the last six years.

Thank you for constantly pushing me towards a success and never giving up.

Thank you for being not just an advisor, but also a true friend.

Thank you for all the inspirational conversations during our coffee breaks and all your encouragement.

And as Socrates said "I cannot teach anybody anything. I can only make them think" thank you Jerry for making me think.

I would also like to thank our technician Zvia Konrad for all her help.

To my lab friends. Thank you for becoming my dearest friends. Thank you for the great time we spent working, laughing, crying and just hanging out together. You made this period a precious time for me, and a joy to come to work every day.

And last but not least, I would like to thank my family and friends for supporting me in all of my decisions.

Abstract

N-glycosylation of proteins is one of the most prevalent post-translational modifications in Nature. However, in contrast to the advanced description of the eukaryal and bacterial pathways of N-glycosylation, the archaeal version of this protein processing event remains the least understood. In the last decade, however, substantial progress has been made in deciphering this process in the halophilic archaea Haloferax volcanii, with the delineation of the archeal glycosylation (Agl) pathway. This pathway is involved in the assembly and the attachment of a pentasaccharide comprising a hexose, two hexuronic acids, a methyl ester of hexuronic acid and a mannose to select Asn residues (Asn-13 and Asn-83) of the surface (S)-layer glycoprotein, a reporter of N-glycosylation in Hfx. volcanii, as well as to Asn residues of other glycoproteins in this organism.

In Hfx. volcanii, it is now known that N-glycosylation involves the addition of nucleotide-activated versions of the first four sugar subunits of the N-linked pentasaccharide to a common dolichol phosphate (DolP) carrier via the ordered actions of four glycosyltransferases (GTases). My work identified AglJ as the first

GTase acting in the Agl pathway, responsible for adding the first hexose to the DolP carrier. Following AglJ, AglG, AglI and AglE sequentially add the next three sugars to the hexose-charged DolP carrier. At the same time, the GTase AglD adds mannose, the final subunit of the pentasaccharide, to a unique DolP carrier. My studies identified catalytic residues of this GTase. Following their assembly, the DolP-bound tetrasaccharide and the DolP-bound mannose are flipped across the plasma membrane to face the exterior. By using computer-based tools together with gene deletion, mass spectrometric analysis of DolP carriers and the S-layer glycoprotein and metabolic radiolabeling, I have shown a role for AglR as a DolP-mannose flippase or

contributing to such activity. After reorientation of the DolP-tetrasaccharide and the

DolP-mannose to face the exterior surface, the tetrasaccharide is transferred to Asn residues by the oligosaccharyltransferase (OST) AglB. Only then is the final sugar, mannose, delivered from its flipped DolP carrier to the N-linked tetrasaccharide.

Although perturbation of N-glycosylation compromises S-layer stability and architecture, as well as S-layer resistance to added proteases, N-glycosylation is not essential for Hfx. volcanii survival. Nonetheless, such post-translational modification offers a response to changing growth conditions. Specifically, the S-layer glycoprotein can be simultaneously modified by two different N-glycans as a function of environmental salinity. S-layer glycoprotein Asn-13 and Asn-83 are modified by the pentasaccharide described above when Hfx. volcanii cells are grown in either 3.4

M or 1.75 M NaCl-containing medium, whereas Asn-498 is modified by a distinct tetrasaccharide comprising a sulfated hexose, two hexoses and rhamnose only when cells are grown at the lower salinity. I have identified components of a novel pathway involved in the assembly of the Asn-498-linked tetrasaccharide. This study represents the first report of two N-glycosylation pathways able to simultaneously modify a single protein in any organism. Still, whereas two N-glycosylation pathways have been described in Hfx. volcanii, only one OST, AglB, has been identified. This observation led me to consider the phylogenetic distribution of N-glycosylation pathways among the archaeal branch of the tree of life, based on the presence or absence of AglB.

Table of Contents

1. Introduction ...... 1

1.1 Post-translational modifications ...... 1

1.2 Protein N-glycosylation in Eukarya ...... 2

1.3 Protein N-glycosylation in Bacteria ...... 4

1.4 Protein N-glycosylation in Archaea ...... 6

1.4.1 N-glycosylation in Haloferax volcanii ...... 8

1.4.2 N-glycosylation in other archaeal model organisms ...... 11

2. Objectives and specific goals ...... 16

3. Results ...... 17

3.1 General overview of papers published during the thesis ...... 17

Chapter 3.2 ...... 24

"AglJ, a novel component of the Haloferax volcanii N-glycosylation pathway"

(Kaminski et al., 2010) ...... 24

Chapter 3.3 ...... 26

"Identification of residues important for the activity of Haloferax volcanii AglD,

a component of the archaeal N-glycosylation pathway (Kaminski and Eichler

2010)………………………………………………………………………..…....26

Chapter 3.4 ...... 27

"AglR is required for addition of the final mannose residue of the N-linked

glycan decorating the Haloferax volcanii S-layer glycoprotein" (Kaminski et al.,

2012)……...... 27

Chapter 3.5 ...... 29

"Two distinct N-glycosylation pathways together process the S-layer

glycoprotein in the halophilic archaea, Haloferax volcanii" (Kaminski et al.,

2013a) ...... 29

Chapter 3.6 ...... 31

"Phylogenetic- and genome-derived insight into the evolution of N-glycosylation

in Archaea" (Kaminski et al., 2013b) ...... 31

4. Discussion ...... 32

4.1 AglJ, the first glycosyltransferase of the Hfx. volcanii Agl pathway ...... 32

4.2 The glycosyltransferase AglD ...... 35

4.3 The archaeal flippase ...... 37

4.4 New Agl pathway ...... 40

4.5 N-glycosylation across the Archaea ...... 44

Appendix A ...... 46

"Distinct glycan-charged phosphodolichol carriers are required for the assembly of

the pentasaccharide N-linked to the Haloferax volcanii S-layer glycoprotein"

(Guan et al., 2010) ...... 46

Appendix B ...... 47

"Protein glycosylation in Archaea: Sweet and Extreme" (Calo et al., 2010) ...... 47

Appendix C ...... 48

"Add salt, add sugar: N-glycosylation in Haloferax volcanii" (Kaminski et al.,

2013c)……………………………………………………………………………..48

5. Contributions to this thesis...... 49

6. Papers resulting from this thesis ...... 50

7. References ...... 52

1. Introduction

1.1 Post-translational modifications

The path a protein follows from its synthesis as a nascent polypeptide chain to the mature functional protein involves a variety of processing events. One major step often involved in the biogenesis of an active protein is chemical modification by any of a number of functional groups. Such post-translational protein modifications are responsible for much of the diversity found within the proteome of any organism. At the same time, post-translational modifications can significantly modulate the physico-chemical and biological properties of a protein through effects on protein function, sub-cellular localization, oligomerization, folding or turnover, as well as determining the participation of a protein in numerous recognition events (for review see Eichler and Adams, 2005; Prabakaran et al., 2012). The changes made to the polypeptide chain by such modifications include the addition of various moieties, such as methyl, acetyl, phosphate, lipid and carbohydrate groups. Of the various post- translational modifications that a protein can undergo, glycosylation, namely the covalent attachment of sugar residues, is the most complex. Indeed, it was reported that glycosylation is likely the most abundant modification affecting eukaryal proteins, with more than two-third of all proteins in eukaryotes predicted to be subject to such processing (Apweiler et al., 1999). The diversity of sugars assembled into glycans, including pentoses, hexoses, hexosamines, deoxyhexoses, uronic acids and sialic acids, as well as the range of sizes such glycans can contain, comprising monosaccharides to large oligosaccharides, coupled with the variety of possible linkages between the components of the glycan, allows for an endless number of possible protein-linked glycan structures.

To date, five major classes of protein glycosylation have been described on the basis of the manner in which the glycan is linked to the modified protein, including N-

, O-, P- and C-glycosylation, as well as glycosylphosphatidylinositol (GPI) anchor addition (Figure 1). In N-glycosylation, the focus of this thesis, select asparagines residue found within a sequon (i.e. a conserved Asn-X-Ser/Thr motif, where X is any amino acid but proline) are covalently modified by an oligosaccharide (Gavel and von

Heijne, 1990).

Figure 1: Schematic representation of the various types of protein glycosylation currently known (Taken from Jarrell et al., submitted).

1.2 Protein N-glycosylation in Eukarya

Protein glycosylation was first reported in the late 1930’s, when Neuberger demonstrated a carbohydrate group to be an integral part of crystalline egg albumin

(Neuberger, 1938). In time, it became generally accepted that protein glycosylation was a trait restricted to Eukarya. In yeast and higher Eukarya, N-glycosylation is a relatively well-understood and evolutionarily-conserved process that begins when soluble activated sugars are sequentially assembled into a N-acetylglucosamine

(GlcNAc)2-mannose (Man)5 heptasaccharide on a dolichol pyrophosphate (Dol-PP) lipid carrier embedded in the cytoplasmic leaflet of the endoplasmic reticulum (ER) membrane (Burda and Aebi, 1999; Spiro, 2002; Helenius and Aebi, 2004). The lipid- linked oligosaccharide (LLO) is then 'flipped' across the membrane to face the ER lumen in an ATP-independent manner. The flippase responsible for this translocation event was initially identified as Rft1 based on yeast genetic studies, however, subsequent biochemical studies questioned this assignment (Helenius et al., 2002;

Frank et al., 2008; Rush et al., 2009). Once oriented towards the ER lumen, the flipped lipid-linked heptasaccharide is further augmented by the sequential attachment of four mannose and three glucose residues, each transferred from individual dolichol phosphate (DolP)-mannose or DolP-glucose units, charged on the cytoplasmic face of the ER membrane and flipped to face the ER lumen by unidentified flippases, distinct from that used to flip the DolPP-heptasaccharide (Sanyal and Menon, 2010). The resulting 14-member branched polysaccharide is transferred by the Stt3 subunit of the multimeric oligosaccharyltransferase (OST) onto the Asn residues of Asn-X-Ser/Thr motifs found in nascent polypeptide chains co-translationally translocating into the

ER lumen (Silberstein and Gilmore, 1996; Yan and Lennarz, 2005), as portrayed in

Figure 2.

The glycoprotein may then pass through the Golgi apparatus before being distributed to different destinations in the cell or secreted from the cell. During this passage, the N-linked oligosaccharide is further modified by the actions of various glycosidases and/or glycosyltransferases (GTases) (Gemmill and Trimble, 1999).

Such downstream modification is a universal feature of the eukaryotic N- glycosylation system.

Figure 2: N-glycosylation pathways in Eukarya. Schematic representation of the N- glycosylation pathway of Saccharomyces cerevisiae. The legend describes the components of the oligosaccharide ultimately added to the target protein (Taken from Szymanski and Wren, 2005).

1.3 Protein N-glycosylation in Bacteria

While N-glycosylation was considered an exclusively eukaryal trait for thirty years since first being reported, evidence supporting the existence of non-eukaryal glycoproteins appeared in the late 1960's, such as the isolation of a phospholipid- glycoprotein complex from the envelope of the Gram-negative bacterium Escherichia coli B (Okuda and Weinbaum, 1968). While these authors reported the carbohydrate- containing moiety to be covalently bound to the protein, they did not demonstrate direct linkage of this group to an amino acid residue in the protein. Later, glycoproteins in the cell envelope of Halobacterium halobium (since renamed

Halobacterium salinarum) were detected both by carbohydrate-detecting periodic acid-Schiff stain and by concanavalin A agglutination (Koncewicz, 1972). It was subsequently shown that a single 194 kDa protein staining positively for the presence

4 of carbohydrates accounted for almost 50% of the Hbt. salinarum cell envelope protein content (Mescher et al., 1974). Then, in 1976, Mescher and Strominger demonstrated that this protein, the surface (S)-layer glycoprotein, was subject to both

N- and O-glycosylation, thereby providing the first example of a non-eukaryotic glycoprotein. Soon after, glycoproteins were detected in the cell walls of the Gram- positive bacteria Clotridium thermosaccharolyticum and Clostridium thermohydrosulfuricum (Sleytr and Thorne, 1976). However it was only years later that the first bacterial N-linked glycosylation system was described in Campylobacter jejuni (Szymanski et al., 1999; Wacker et al., 2002).

Today, with the identification of genes involved in this post-translational modification in C. jejuni, this system remains the best-studied case of N-linked glycosylation in bacteria (Nothaft and Szymanski, 2010). Here, the products of the pgl locus act to assemble a branched heptasaccharide on a lipid-linked carrier, undecaprenol pyrophosphate (UndPP) (Linton et al., 2005). The UndPP- heptasaccharide is subsequently flipped across the cytoplasmic membrane to the periplasmic side by an ATP-dependent flippase, PglK (formerly named WlaB), although N-glycosylation persists in a strain lacking pglK (Alaimo et al., 2006). The polysaccharide is then transferred to selected Asn residues located within an expanded sequon by PglB, a monomeric OST homologous to the Stt3 subunit of the eukaryotic

OST complex (Feldman et al., 2005), as reflected in Figure 3. In contrast to the eukaryal N-glycosylation process, no further modifications of the N-linked heptasaccharide occurs in C. jejuni.

Figure 3: N-glycosylation pathway in Bacteria. Schematic representation of the N- glycosylation pathway in Campylobacter jejuni. The legend describes the components of the heptasaccharide added to target proteins (Taken from Szymanski and Wren, 2005).

1.4 Protein N-glycosylation in Archaea

While a substantial body of work has addressed the N-glycosylation pathway of

C. jejuni (Szymanski et al., 1999; Szymanski and Wren, 2005; Wacker et al., 2002;

Nothaft and Szymanski, 2010), it was the surface (S)-layer glycoprotein of the halophilic archaeon Halobacterium salinarum (Mescher and Strominger, 1976) that provided the first example of non-eukaryotic protein glycosylation, challenging the dogma that only eukaryal proteins are subject to such post-translational modification.

In 1976, the year that Hbt. salinarum was shown to contain true glycoproteins, this organism was still considered a member of the bacterial world, albeit an unusual one.

However, following Carl Woese's pioneering use of 16S/18S ribosomal (r)RNA analysis reported the next year (Fox et al., 1977; Woese and Fox, 1977), an approach that ultimately led to the redrawing of the universal tree of life to comprise three

6 distinct domains, i.e. Eukarya, Bacteria and Archaea, Hbt. salinarum was reassigned to the archaeal branch and the field of archaeal protein glycosylation was founded.

Although genetic tools were not available at the time when the Hbt. salinarum

S-layer glycoprotein was first shown to be N-glycosylated (i.e. 1976), structural and biochemical approaches were successfully employed to reveal various aspects of the

N-glycosylation pathway in this organism. In Hbt. salinarum, the S-layer glycoprotein forms a rigid structural matrix at the cell surface that is considered to be responsible for the rod-shaped morphology of this organism (Mescher and Strominger, 1977).

Detailed analysis showed this protein to be modified by two different Asn-linked oligosaccharides, namely a repeating sulfated pentasaccharide linked to Asn-2 via N- acetylgalactosamine and a sulfated glycan linked to ten other Asn residues through a glucose residue (Mescher and Strominger, 1978; Wieland et al., 1980; Wieland et al.,

1983; Lechner et al., 1985a; Lechner and Wieland, 1989). Moreover, it was shown that while DolP serves as the lipid glycan carrier during assembly of the glucose- linked sulfated glycan, DolPP bears the repeating sulfated pentasaccharide N- acetylgalactosamine linked to Asn-2 of the S-layer glycoprotein (Wieland et al., 1980;

Lechner and Wieland, 1989). In addition, the glucose-linked sulfated glycan was shown to be methylated in the DolP-linked form but not when protein-bound, indicating that methylation is an obligatory step during glycoprotein synthesis, although the role of this modification remains unclear (Lechner et al., 1985b). Finally, studies showing the ability of Hbt. salinarum cells to modify cell-impermeable, sequon-bearing hexapeptides with sulfated oligosaccharides served to localize the N- glycosylation event to the external cell surface (Lechner et al., 1985a).

Not long after the Hbt. salinarum S-layer glycoprotein provided the first example of N-glycosylation in Archaea, other experimentally confirmed examples of

7 similarly modified polypeptides were described in this and other archaeal species.

These included the Hbt. salinarum flagellin (recently renamed the archaellin (Jarrell and Albers, 2012)), shown to be glycosylated by the same glycan-linked sulfated polysaccharide as the S-layer glycoprotein in this organism (Wieland et al., 1985), as well as S-layer glycoproteins from a second haloarchaeon Haloferax volcanii. In Hfx. volcanii, it was reported that Asn-13 and Asn-498 of the S-layer glycoprotein are modified by a linear string of glucose residues, whereas Asn-274 and/or Asn-279 were supposedly decorated by a glycan containing glucose, galactose, and idose subunits (Sumper et al., 1990). Still, despite these initial advances, detailed understanding of the archaeal N-glycosylation process had to wait for the development of appropriate molecular tools.

1.4.1 N-glycosylation in Haloferax volcanii

Despite early structural and biochemical advances in deciphering the pathway of N-glycosylation in Hbt. salinarum, the N-glycosylation process is currently best understood in Hfx. volcanii (Sumper et al., 1990). Specifically, substantial progress has been made in deciphering the process of N-glycosylation in Hfx. volcanii in the past decade with the delineation of the Agl (archaeal glycosylation) pathway in this organism. At the beginning of my doctoral research, it has been shown by previous studies conducted in the Eichler laboratory that the Hfx. volcanii S-layer glycoprotein is modified at Asn-13 and Asn-83 by a pentasaccharide containing two hexoses, two hexuronic acids and a 190 Da sugar residue. agl genes involved in the assembly and attachment of this N-linked glycan were originally identified on the basis of the homology of their predicted protein products to components of well-defined eukaryal or bacterial N-glycosylation pathways (Abu-Qarn and Eichler, 2006). After verifying

8 the transcription of the identified Hfx. volcanii genes and following gene deletions based on the "pop-in/pop-out" approach developed for this organism (Allers et al.,

2004), the roles of Agl proteins in N-glycosylation of the S-layer glycoprotein were verified by following the migration of the S-layer glycoprotein from deletion strains on SDS-PAGE followed by periodic acid-Schiff reagent or Coomassie blue staining, or by matrix-assisted laser desorption/ionization time of flight (MALDI TOF) mass spectrometry analysis of S-layer glycoprotein-derived peptides containing Asn-13 and

Asn-83 generated upon digestion of the surface-layer glycoprotein with Glu-C protease and/or trypsin (Abu-Qarn and Eichler, 2006; Abu-Qarn et al., 2007; Abu-

Qarn et al., 2008; Yurist-Doutsch et al., 2008). Later, in an effort to identify novel components of the Hfx. volcanii N-glycosylation process not identified through such homology-based searches, open reading frames (ORFs) found adjacent to known agl genes were considered. Computer-based approaches and visual inspection of the Hfx. volcanii genome thus identified novel agl genes and defined an agl gene cluster spanning an 18 kbp region of the genome starting at HVO_1517 (aglJ) and ending with HVO_1530 (aglB) that included all but one of the genes implicated in the Hfx. volcanii N-glycosylation process identified to that point (Yurist-Doutsch and Eichler,

2009).

At the same time, general functions were assigned to the various Agl proteins.

AglG participates in the addition of the hexuronic acid found at position two of the pentasaccharide, while AglI and AglF are involved in the addition of the hexuronic acid detected at position three of the glycan (Yurist-Doutsch et al., 2008). AglM is involved in the biosynthesis of the hexuronic acid found at position two of the pentasaccharide, acting as a UDP-glucose dehydrogenase that is able to utilize various sugar nucleotide substrates. It was further demonstrated that AglM is able to act in

9 concert with AglF, shown to be a glucose-1-phosphate uridyltransferase, implying that AglM also participates in the biogenesis of the hexuronic acid found at pentasaccharide position three (Yurist-Doutsch et al., 2010). AglE was found to participate in the addition of the 190 Da sugar residue detected at pentasaccharide position four (Abu-Qarn et al., 2008). AglP was shown to function as a S-adenosyl-L- methionine (SAM)-dependent methyltransferase responsible for methylation of the hexuronic acid found at pentasaccharide position four (Magidovich et al., 2010).

AglD, encoded by the only N-glycosylation gene not found in the agl gene cluster, contributes to the addition of the final hexose of the pentasaccharide (Abu-Qarn et al.,

2007). Finally, AglB acts as the OST in Hfx. volcanii, delivering the assembled pentasaccharide to at least two target Asn residues in the S-layer glycoprotein (Abu-

Qarn et al., 2007). A model of the Hfx. volcanii Agl pathway, based on what was known until 2010, as described above, is presented in Figure 4.

Figure 4: Schematic representation of the N-glycosylation pathway in Hfx. volcanii, circa 2010. AglG, AglI, AglE, AglD, AglF, AglM and AglP are involved in the assembly of a pentasaccharide that is later transferred to select Asn residues on the S-layer glycoprotein by the OST AglB. The components of the pentasaccharide are listed.

Thus, while progress had been made in characterizing the N-glycosylation process in Haloferax volcanii, many questions still remained unanswered.

1.4.2 N-glycosylation in other archaeal model organisms

While the archaeal N-glycosylation process is currently best understood in Hfx. volcanii, insight into archaeal N-glycosylation has also come from other species, specifically from the methanogens Methanococcus voltae and Methanococcus maripaludis and the thermoacidophile Sulfolobus acidocaldarius.

The first archaeal genes involved in N-glycosylation were reported in

Methanococcus voltae PS. In this strain, glycoproteins are modified by a N-linked trisaccharide consisting of GlcNAc, di-acetylated glucuronic acid and acetylated mannuronic acid that is modified by the attachment of a threonine (Voisin et al.,

2005) or by the same trisaccharide extended by an additional 220 kDa or 262 kDa residue that was identified from a different isolate of the same strain (Chaban et al.,

2009). Genetic analysis together with mass spectrometry suggest, as in other Archaea, that AglB is the OST that transfers the lipid-bound glycan to selected Asn residues in the target glycoprotein (Chaban et al., 2006). Such analysis identified AglH as adding

GlcNAc, the first sugar of the N-linked glycan to DolP. The role of AglH is supported by the ability of the aglH gene to complement a conditionally lethal mutation in S. cerevisiae alg7, the gene encoding the enzyme responsible for adding GlcNAc to the

DolP carrier in the eukaryotic N-glycosylation pathway (Shams-Eldin et al., 2008).

The GTases AglC and AglK are involved in the addition of the second sugar, a diacetylated glucuronic acid, while AglA is involved in the addition of acetylated mannuronic acid, the final sugar of the glycan (Chaban et al., 2006). Recently, in vitro studies suggested that AglK, rather than AglH, is the glycosyltransferase needed for

11 the first step of the N-glycosylation process (Larkin et al., 2013). Current models for

N-glycosylation in M. voltae are presented in Figure 5.

Figure 5: N-glycosylation pathway in M. voltae. Schematic representation of the N- glycosylation pathway in M. voltae according to genetics work from the Jarrell group (top panel) and according to the in vitro assay of Larkin et al. (bottom panel). The legend describes the components of the oligosaccharide (Taken from Eichler, 2013 and Larkin et al., 2013, respectively).

In M. maripaludis S2, a tetrasaccharide slightly different from the trisaccharide attached to proteins of M. voltae and composed of N-acetylgalactosamine (GalNAc) as a linking sugar, di-acetylated glucuronic acid as the second sugar, acetylated mannuronic acid that is modified by the attachment of a threonine and an additional acetamidino group added at position C-3 and a fourth sugar, (5S)-2-acetamido-2,4- dideoxy-5-O-methyl-α-L-erythro-hexos-5-ulo-1,5-pyranose (Sug), was found to be N- linked to archaellins. Here as well, the glycan is transferred by the OST AglB

(VanDyke et al., 2009). The pathway begins with the addition of UDP-GalNAc to

DolP in an as yet unknown manner. The next three nucleotide-activated sugars are added by the AglO, AglA and AglL GTases, respectively (VanDyke et al., 2009). The pathway for formation of the UDP-activated second sugar starts with fructose-6- phosphate (Fru-6-P) and involves multiple enzymes, including MMP1680,

MMP1077, MMP1076, MMP0352, MMP0352, MMP0351 and MMP0350 (Namboori and Graham, 2008). The third sugar is generated by MMP0357 and is further modified by AglX, AglY and AglZ that act to add an acetamidino functional group

(Jones et al., 2012). The final sugar is added by AglL and modified with a methyl group by AglV, at which point the third sugar is also modified with a threonine by

AglU (Ding et al., 2013). The current model of M. maripaludis N-glycosylation is presented in Figure 6.

Figure 6: N-glycosylation pathway in M. maripaludis. Schematic representation of the N-glycosylation pathway in M. maripaludis. (Taken from Jarrell et al., submitted).

With a growing list of tools now available for manipulation of the thermoacidophile Sulfolobus acidocaldarius, the pathway responsible for the biosynthesis of the N-linked tri-branched hexasaccharide decorating glycoproteins in this species is emerging. The N-glycan of S. acidocaldarius is a tri-branched

13 hexasaccharide composed of [Hex]4HexNAc(6-sulfo-Qui)-HexNAc, where the four hexoses are mannose and glucose, while sulpho-Qui is a rare sulfated sugar, sulfoquinovose (Zähringer et al., 2000), previously found in the photosynthetic membranes of all higher plants, mosses, ferns and algae, as well as in most photosynthetic bacteria (Benning, 1998). The glycan of S. acidocaldarius is apparently assembled on a short and highly saturated lipid carrier on the cytoplasmic side of the cell membrane (Peyfoon et al., 2010; Guan et al., 2011). At present, it is not clear whether the N-glycan is assembled on a DolP or DolPP lipid carrier. The first sugar residue, GlcNAc, is transferred to the lipid carrier by AglH. In the next step, Saci1262 is proposed to transfer GlcNAc to create a lipid-linked chitobiose

(GlcNAc(β1-4)GlcNAc) core. This same chitobiose core is conserved in eukaryal N- linked glycans (Burda and Aebi, 1999). In the third step of this process, one of the terminal mannose residues is transferred onto the chitobiose core, creating either the α

1-4 or α 1-6 linkage to the second GlcNAc. The fourth step involves Agl3 that converts UDP-glucose and sodium sulfite into UDP-sulfoquinovose, which is then transferred onto the lipid-linked trisaccharide in an as yet unknown manner (Meyer et al., 2011). In the final step, the soluble glycosyltransferase Agl16 transfers the terminal mannose and glucose residues (Meyer et al., 2013). As in other Archaea where N-glycosylation has been studied, the glycan is transferred to selected Asn residues of a target protein by the OST AglB. Also as elsewhere, the identity of the enzyme responsible for the flipping of the lipid-linked glycan from one side of the membrane to the other prior to delivery of the glycan to the target protein is unknown.

The current model of S. acidocaldarius N-glycosylation is presented in Figure 7.

Figure 7: N-glycosylation pathway in S. acidocaldarius. Schematic representation of the N-glycosylation pathway of S. acidocaldarius. (Taken from Jarrell et al., submitted).

2. Objectives and specific goals

The goal of my research was to further describe N-glycosylation in Hfx. volcanii by identifying additional components of the pathway, as well as to biochemically characterize enzymes that contribute to this process.

Specifically, my aims were to:

1. Identify the glycosyltransferase involved in the assembly of the first sugar

residue of the pentasaccharide N-linked to the S-layer glycoprotein

2. Biochemically characterize AglD, the glycosyltransferase involved in the

assembly of the last sugar residue of the pentasaccharide

3. Identify the archaeal flippase

4. Examine the N-glycosylation process at different salt concentrations

5. Obtain insight into the distribution and evolutionary history of the archaeal N-

glycosylation process

3. Results

3.1 General overview of papers published during the thesis

The results obtained in this study are summarized in the following chapters that include five published articles (Chapters 3.2-3.6). The overall conclusions of these studies are summarized and discussed in Chapter 4. In addition, another published article and two review articles that emerged from our laboratory during my Ph.D. studies and to which I contributed appear in the Appendix.

In my first article (Chapter 3.2), entitled "AglJ, a novel component of the

Haloferax volcanii N-glycosylation pathway" (Kaminski, L., Naparstek, S., Abu-

Qarn, M., Eilam, Y., Guan, Z., Raetz, C.R., Lichtenstein, R. and Eichler, J. 2010. J.

Bacteriol. 192:5572-5579), the participation of HVO_1517 in the N-glycosylation process in Hfx. volcanii was shown. HVO_1517 was first identified through its homology to Dpm1, a known component of the eukaryal N-glycosylation pathway

(Abu-Qarn and Eichler, 2006). In addition, HVO_1517 is found in genomic proximity to agl genes encoding known elements of the Hfx. volcanii N-glycosylation process

(Yurist-Doutsch and Eichler, 2009). By combining deletion of HVO_1517 with mass spectrometric analysis of both DolP-monosaccharide-charged carriers and the S-layer glycoprotein, evidence was obtained showing the participation of HVO_1517, renamed AglJ, in adding the first hexose of the N-linked pentasaccharide decorating the S-layer glycoprotein. Since the deletion of aglJ did not fully prevent the attachment of a hexose residue to the S-layer glycoprotein, to better understand the origin of the minor monosaccharide-modified peaks associated with both the S-layer glycoprotein-derived peptide and the DolP carrier in the absence of AglJ, I examined

17 the monosaccharide-charged DolP population. Such analysis revealed that AglJ is involved in modifying only one of the three Hfx. volcanii monosaccharide-charged dolichol phosphates detected. Nonetheless, in cells lacking AglJ, no further sugar subunits were added to the remaining monosaccharide-charged DolP carriers or to the monosaccharide-modified S-layer glycoprotein, pointing to the importance of the sugar added through the actions of AglJ for proper N-glycosylation assembly. It was also shown that HVO_1613, another Hfx. volcanii Dpm1 homologue also generates monosaccharide-modified dolichol phosphate although this entity does not contribute to N-glycosylation.

With the identification of the glycosyltransferases involved in the assembly of the pentasaccharide N-linked to the S-layer glycoprotein, namely, AglJ, AglG, AglI,

AglE and AglD (Abu-Qarn et al., 2007; Abu-Qarn et al., 2008; Yurist-Doutsch et al.,

2008; Kaminski et al., 2010), responsible for the addition of the first, second, third, fourth and final sugars, respectively, efforts were next focused on biochemical analysis. My second article (Chapter 3.3), entitled "Identification of residues important for the activity of Haloferax volcanii AglD, a component of the archaeal N- glycosylation pathway" (Kaminski, L. and Eichler, J. 2010. Archaea 6;2010:315108), focused on biochemical analysis of the glycosyltransferase AglD. In this study, I designed an in vivo assay to identify amino acids important for AglD activity. In this assay, restoration of AglD function in a Hfx. volcanii aglD deletion strain was tested upon transformation with plasmid-encoded versions of AglD mutants generated through site-directed mutagenesis and following the migration and periodic acid-

Schiff staining of the S-layer glycoprotein in each strain. Mutations were introduced at positions corresponding to conserved residues in archaeal homologues of AglD.

This assay, together with homology modeling of AglD based on the three dimensional

18 structure of a glycosyltransferase from Bacillus subtilis, assigned AglD Asp110 and

Asp112 as elements of the DXD motif. This motif interacts with metal cations associated with nucleotide-activated sugar donors, while Asp201 was predicted to be the catalytic base of the enzyme.

At this point, it was determined that the N-linked pentasaccharide decorating the

Hfx. volcanii S-layer glycoprotein is assembled on two different DolP carriers, i.e. the first four sugars are assembled on one DolP carrier while the last sugar mannose is assembled on its own DolP carrier (Guan et al., 2010). Both sugar-charged DolPs are assembled on the cytoplasmic face of the plasma membrane (Plavner and Eichler,

2008). However, the sugars are transferred to the S-layer glycoprotein on the external surface of the cell. As such, the flippase responsible for delivering tetrasaccharide- and/or monosaccharide-charged DolP carriers across the plasma membrane, as well as the enzyme responsible for delivering mannose from its DolP carrier to the protein- bound tetrasaccharide were next sought. My third article (Chapter 3.4), entitled "AglR is required for addition of the final mannose residue of the N-linked glycan decorating the Haloferax volcanii S-layer glycoprotein" (Kaminski, L., Guan, Z., Abu-Qarn, M.,

Konrad, Z. and Eichler, J. 2012. Biochim. Biophys. Acta. 1820:1664-1670), describes the identification of AglR as a DolP-mannose flippase. AglR was initially identified as part of the agl gene cluster and was shown to be co-transcribed with aglE (Yurist-

Doutsch and Eichler, 2009). The homology of AglR to Wzx, a bacterial protein thought to translocate lipid-linked O-antigen precursor oligosaccharides across the plasma membrane (Liu et al., 1996), justified considering AglR as a flippase. Deletion of aglR affected S-layer glycoprotein migration on SDS-PAGE. Next, mass spectrometry analysis of the glycan-charged DolP pool from the AglR-lacking strain showed an accumulation of tetrasaccharide-charged DolP, as well as of its

19 trisaccharide-charged precursor. In addition, the level of AglD-generated DolP- mannose was higher in aglR cells than in the parent strain. Moreover, the N- glycosylation profile of a S-layer glycoprotein-derived Asn-13-containing peptide from the deletion strain revealed modification by only the first four subunits of the pentasaccharide and not by the complete pentasaccharide. This, together with the observation that 2-[3H]-mannose was not incorporated into the S-layer glycoprotein in the ∆aglR strain, led me to assign AglR as either serving as the Dol-P-mannose flippase or contributing to such activity, as well as affecting the flipping of tetrasaccharide-charged DolP.

Until recently, all of the experiments performed in our group involved Hfx. volcanii cells grown in the presence of 3.4 M NaCl-containing medium. Since Hfx. volcanii was originally reported to grow in medium containing NaCl ranging from 1-4

M (Mullakhanbhai and Larsen, 1975), the N-glycosylation profile of the S-layer glycoprotein from cells grown at low salinity (1.75 M NaCl-containing medium) was recently examined in our group (Guan et al., 2012). When grown at the lower salinity, the Asn-13 and Asn-83 positions were still modified by the pentasaccharide discussed above, albeit to a lesser degree. More strikingly, S-layer glycoprotein Asn-498, a position not modified when cells are grown in the presence of high salt, was modified by a tetrasaccharide of different composition from the pentasaccharide decorating

Asn-13 and Asn-83 (Guan et al., 2012). Thus, in my fourth publication (Chapter 3.5), entitled "Two distinct N-glycosylation pathways together process the S-layer glycoprotein in the halophilic archaea, Haloferax volcanii" (Kaminski, L., Guan, Z.,

Yurist-Doutsch, S. and Eichler, J. 2013a. mBio. 4:e00716-13), I examined the N- glycosylation process at different salt concentrations. To determine whether the same

Agl pathway that is responsible for the assembly and attachment of the N-linked

21 pentasaccharide is also responsible for generating the N-linked tetrasaccharide seen when cells are grown under low salt conditions, the N-glycosylation profile of the S- layer glycoprotein in cells lacking genes belonging to the Agl pathway i.e. in aglI and aglE cells, was examined by mass spectrometry analysis at both the glycan- charged DolP and S-layer glycoprotein levels. While the absence of these genes compromised pentasaccharide assembly as expected, such deletions had no effect on the appearance of the ‘low salt’ tetrasaccharide attached to S-layer glycoprotein Asn-

498 at low salinity. To identify genes involved in the biogenesis of the ‘low salt’ tetrasaccharide, I scanned the Hfx. volcanii genome for a cluster of genes currently annotated as serving sugar processing-related roles. One such cluster was HVO_2046 to HVO_2061. The rationale was that if all but one of the known agl genes are organized into a single gene cluster (Yurist-Doutsch and Eichler, 2009), the same may hold true for genes of the novel N-glycosylation pathway. I performed deletion of each gene within this cluster as previously described (Allers et al., 2004). Each deleted strain grown in low salt medium was subjected to mass spectrometry analysis of both the glycan-charged DolP and S-layer glycoprotein pools, which confirmed that components of this gene cluster, renamed agl5-agl15, are involved in the biosynthesis of the low salt tetrasaccharide N-linked to S-layer glycoprotein Asn-498.

In addition, I was able to show interplay between the two Hfx. volcanii N- glycosylation pathways. While the low-salt tetrasaccharide is barely detectable when

Hfx. volcanii cells are grown in the higher salinity, this glycan was readily detected in strains lacking pentasaccharide biosynthesis pathway genes, under such high salt growth conditions. When a strain lacking AglB, the only known OST in Hfx. volcanii or indeed in any other Archaea, was subjected to growth in low salt medium and addressed by mass spectrometry, it was revealed that the transfer of the low salt

21 tetrasaccharide from the DolP carrier upon which it is assembled to S-layer glycoprotein Asn-498 persisted AglB. At this point is not known how this transfer is realized.

The observation that Hfx. volcanii encodes two separate pathways for N- glycosylation but that only one OST AglB is found in this organism was the basis for my fifth paper (Chapter 3.6), entitled "Phylogenetic- and genome-derived insight into the evolution of N-glycosylation in Archaea" (Kaminski, L., Lurie-Weinberger, MN.,

Allers, T., Gophna, U and Eichler, J. 2013b. Mol. Phylogenet. Evol. 68:327-339). To gain insight into the distribution and evolutionary history of the archaeal version of N- glycosylation, I used bioinformatics tools to scan 168 archaeal genome sequences for the presence of aglB. Of these 168 available archaeal genomes, the presence of AglB is predicted in 166 species. While some species encode a single version of the protein, others contain multiple versions. Phylogenetic analysis revealed that the events leading to aglB duplication occurred at various points during archaeal evolution. In many cases, aglB is found as part of a gene cluster of putative N-glycosylation-related genes. When assessing the phylogenetic relationships and the presence, arrangement and nucleotide composition of genes in aglB-based clusters from five species of

Haloferax i.e. Hfx. volcanii, Hfx. denitrificans, Hfx. mediterranei, Hfx. mucosum and

Hfx. sulfurifontis, lateral gene transfer was deemed as contributing to the evolution of archaeal N-glycosylation.

In addition to these publications as first author, I have also contributed to other papers from the Eichler laboratory that appear in the Appendix of this thesis.

Appendix A, entitled "Distinct glycan-charged phosphodolichol carriers are required for the assembly of the pentasaccharide N-linked to the Haloferax volcanii S- layer glycoprotein" (Guan, Z., Naparstek, S., Kaminski, L., Konrad, Z. and Eichler, J.

2010. Mol. Microbiol. 78:1294-1303), describes that the pentasaccharide decorating

S-layer glycoprotein Asn-13 and/or Asn-83 is assembled on two different DolP carriers. My contribution to this paper was the isolation of a total lipid extract from

Hfx. volcanii cells lacking AglD for the mass spectrometry analysis.

Appendix B, entitled "Protein glycosylation in Archaea: Sweet and Extreme"

(Calo, D., Kaminski, L. and Eichler, J. 2010. Glycobiology 20:1065-1076), is a review article describing N-glycosylation in Archaea. I helped to collect relevant information from the literature and contributed to the writing of the paper.

Appendix C, entitled "Add salt, add sugar: N-glycosylation in Haloferax volcanii" (Kaminski, L., Naparstek, S., Kandiba, L., Cohen-Rosenzweig, C., Arbiv,

A., Konrad, Z. and Eichler, J. 2013c. Biochem. Soc. Trans. 41:432-435) is a mini- review article describing the N-glycosylation pathway of Haloferax volcanii.

In addition, I have contributed to another review article that was recently submitted to Microbiology and Molecular Biology Reviews entitled "N-linked glycosylation in Archaea: A structural, functional and genetic analysis".

Chapter 3.2

"AglJ, a novel component of the Haloferax volcanii

N-glycosylation pathway"

Lina Kaminski, Mehtap Abu-Qarn, Ziqiang Guan, Shai Naparstek, Valeria V. Ventura, Christian R. H. Raetz, Paul G. Hitchen, Anne Dell and Jerry Eichler. J. Bacteriol. 2010. 192:5572-5579

Abstract:

Like the Eukarya and Bacteria, the Archaea also perform N-glycosylation.

Using the haloarchaeon Haloferax volcanii as a model system, a series of Agl proteins involved in the archaeal version of this posttranslational modification has been identified. In the present study, the participation of HVO_1517 in N-glycosylation was considered, given its homology to a known component of the eukaryal N- glycosylation pathway and because of the genomic proximity of HVO_1517 to agl genes encoding known elements of the Hfx. volcanii N-glycosylation process. By combining the deletion of HVO_1517 with mass spectrometric analysis of both dolichol phosphate monosaccharide-charged carriers and the S-layer glycoprotein, evidence was obtained showing the participation of HVO_1517, renamed AglJ, in adding the first hexose of the N-linked pentasaccharide decorating this reporter glycoprotein. The deletion of aglJ, however, did not fully prevent the attachment of a hexose residue to the S-layer glycoprotein. Moreover, in the absence of AglJ, the level of only one of the three monosaccharide-charged dolichol phosphate carriers detected in the cell was reduced. Nonetheless, in cells lacking AglJ, no further sugar subunits were added to the remaining monosaccharide-charged dolichol phosphate carriers or to the monosaccharide-modified S-layer glycoprotein, pointing to the importance of the sugar added through the actions of AglJ for proper N-glycosylation. Finally, while

24 aglJ can be deleted, Hfx. volcanii surface layer integrity is compromised in the absence of the encoded protein.

AglJ Adds the First Sugar of the N-Linked Pentasaccharide Decorating the Haloferax volcanii S-Layer Glycoproteinᰔ Lina Kaminski,1 Mehtap Abu-Qarn,1 Ziqiang Guan,2 Shai Naparstek,1 Valeria V. Ventura,3 Christian R. H. Raetz,2 Paul G. Hitchen,3,4 Anne Dell,3 and Jerry Eichler1* Department of Life Sciences, Ben Gurion University, Beersheva 84105, Israel1; Department of Biochemistry, Duke University Medical Center, Durham, North Carolina 277102; and Division of Molecular Biosciences,3 and Centre for Integrative Systems Biology,4 Faculty of Natural Sciences, Imperial College, London SW7 2AZ, United Kingdom

Received 18 June 2010/Accepted 14 August 2010

Like the Eukarya and Bacteria, the Archaea also perform N glycosylation. Using the haloarchaeon Haloferax volcanii as a model system, a series of Agl proteins involved in the archaeal version of this posttranslational modification has been identified. In the present study, the participation of HVO_1517 in N glycosylation was considered, given its homology to a known component of the eukaryal N-glycosylation pathway and because of the genomic proximity of HVO_1517 to agl genes encoding known elements of the H. volcanii N-glycosylation process. By combining the deletion of HVO_1517 with mass spectrometric analysis of both dolichol phosphate monosaccharide-charged carriers and the S-layer glycoprotein, evidence was obtained showing the participation of HVO_1517, renamed AglJ, in adding the first hexose of the N-linked pentasaccharide decorating this reporter glycoprotein. The deletion of aglJ, however, did not fully prevent the attachment of a hexose residue to the S-layer glycoprotein. Moreover, in the absence of AglJ, the level of only one of the three monosaccharide-charged dolichol phosphate carriers detected in the cell was reduced. Nonetheless, in cells lacking AglJ, no further sugar subunits were added to the remaining monosaccharide-charged dolichol phosphate carriers or to the monosaccharide-modified S-layer glycoprotein, pointing to the importance of the sugar added through the actions of AglJ for proper N glycosylation. Finally, while aglJ can be deleted, H. volcanii surface layer integrity is compromised in the absence of the encoded protein.

N glycosylation is a posttranslational modification of pro- HVO_1517, encoding one of the five previously identified H. teins in all three domains of life. However, in contrast to our volcanii homologues of eukaryal dolichol phosphate man- relatively advanced description of the eukaryal and bacterial nosyltransferase 1 (Dpm1-C) (1), responsible for catalyzing N-glycosylation pathways (12, 24, 26), many questions re- the transfer of mannose from GDP-mannose to dolichol garding the parallel process in the Archaea remain. With the phosphate in the endoplasmic reticulum membrane (5), was identification of a series of agl (archaeal glycosylation) genes considered here. Indeed, the genomic proximity of in the haloarchaeon Haloferax volcanii and the methanogens HVO_1517 to other genes involved in H. volcanii N glyco- Methanococcus voltae and Methanococcus maripaludis, some sylation warrants such an analysis (27). Relying on mass insight into archaeal N glycosylation is, however, available spectrometry (MS) analysis of an HVO_1517 deletion strain, (for a review, see references 6 and 28). For H. volcanii, the contribution of the encoded protein, renamed AglJ, to N AglB, AglD, AglE, AglF, AglG, AglI, AglM, and AglP were glycosylation is shown. The results demonstrate the partic- shown to participate in the assembly and attachment of a ipation of AglJ in adding the first hexose of the N-linked pentasaccharide to select Asn residues of the surface (S)- pentasaccharide decorating the H. volcanii S-layer glycop- layer protein, a reporter of N glycosylation in this species rotein, serving to generate a monosaccharide-modified doli- (23). Specifically, AglG, AglI, AglE, and AglD are thought chol phosphate carrier. Finally, this study reveals the impor- to be glycosyltransferases involved in adding the second, tance of AglJ action for the proper assembly of the H. third, fourth, and fifth pentasaccharide subunits (2, 3, 29), volcanii surface layer, formed from the S-layer glycoprotein. respectively; AglF is a glucose-1-phosphate uridyltransferase (30); AglM is a UDP-glucose dehydrogenase (30); MATERIALS AND METHODS AglP is an S-adenosyl-L-methionine-dependent methyltransferase (17); and AglB is the oligosaccharyltransferase (2). Strains and growth conditions. H. volcanii parent strain WR536 (H53) and the Despite these advances, proteins catalyzing several cen- isogenic strain deleted of HVO_1517 were grown in medium containing 3.4 M ⅐ tral steps in the H. volcanii N-glycosylation pathway have yet NaCl, 0.15 M MgSO4 7H2O, 1 mM MnCl2, 4 mM KCl, 3 mM CaCl2, 0.3% (wt/vol) yeast extract, 0.5% (wt/vol) tryptone, and 50 mM Tris-HCl (pH 7.2) at to be described. Accordingly, one such candidate, 40°C (19). Deletion of HVO_1517. The deletion of H. volcanii HVO_1517 was achieved by using a previously described approach (1, 4). To amplify approximately 500-bp-long * Corresponding author. Mailing address: Dept. of Life Sciences, regions flanking the coding sequence of HVO_1517, primers HVO_1517-5Јupfor Ben Gurion University, P.O. Box 653, Beersheva 84105, Israel. Phone: (gggctcgagGCTCGCGCAACTCATCAGAG [the genomic sequence is in capital (972) 8646 1343. Fax: (972) 8647 9175. E-mail: [email protected]. letters]) and HVO_1517-5Јuprev (cccaagcttGTCTCGATATGATTGTGCTG), di- ᰔ Published ahead of print on 27 August 2010. rected against the upstream flanking region, and primers HVO_1517-3Јdownfor

5572 VOL. 192, 2010 HALOFERAX VOLCANII AglJ 5573

(gggggatccCGAACCGGGCCGTCGGGACTC) and HVO_1517-3Јdownrev (ccctc tagaCGACCGTGGCACCGGGCGAGG), directed against the downstream flanking region, were employed. XhoI and HindIII sites were introduced by using primers HVO_1517-5Јupfor and HVO_1517-5Јuprev, respectively, while BamHI and XbaI sites were introduced by using primers HVO_1517-3Јdownfor and HVO_1517- 3Јdownrev, respectively. To confirm the deletion of HVO_1517 at the DNA level, PCR amplification was performed by using forward primers directed against an internal region of either HVO_1517 (HVO_1517-for [ATGCCCACCCCCGATGCCGTC]) or trpA (cccgaattcTTATGTGCGTTCCGGATGCG) together with a reverse primer against a region downstream of HVO_1517 (HVO_1517-5Јdownrev), respectively, yielding primer pairs a and b, or by using primers HVO_1517-for and HVO_1517-rev (TCACTCCAGTTCTTCGATTC), designed to amplify a section of the HVO_1517 coding region (primer pair c). Reverse transcription (RT)-PCR was performed as described previously (1), using primer pair c to test for HVO_1517 transcription so as to confirm the HVO_1517 deletion at the RNA level. Isolation of the H. volcanii lipid fraction. The total lipid contents from H. volcanii cells were extracted as follows. Cells were harvested (8,000 ϫ g for 30 min at 4°C) and frozen at Ϫ20°C until extraction was performed. At that point, the pelleted cells were thawed, resuspended in double-distilled water (DDW) (1.33 ml DDW/g cells) and DNase (1.7 ␮g/ml; Sigma, St. Louis, MO), and stirred overnight at room temperature. Methanol and chloroform were added to the cell extract to yield a methanol-to-chloroform-to-cell extract FIG. 1. Deletion of HVO_1517 does not affect cell viability but ratio of 2:1:0.8. After stirring for 24 h at room temperature, the mixture was enhances S-layer glycoprotein SDS-PAGE migration. (A, left and centrifuged (1,075 ϫ g for 30 min at 4°C). The supernatant fractions were middle) PCR amplification was performed by using a forward collected, combined, and filtered through glass wool. Chloroform and DDW primer directed at the 5Ј HVO_1517 flanking region and a reverse were added to the filtrate to yield a chloroform-to-DDW-to-filtrate ratio of primer directed at a sequence within the HVO_1517 coding region 1:1:3.8 in a separating funnel. After separation, the lower clear organic phase, (yielding primer pair a) or using a forward primer directed at a containing the total lipid extract, was collected into a round-bottomed flask sequence within the trpA sequence and the same reverse primer as and evaporated in a rotary evaporator at 35°C. For analysis of the dolichol that described above (yielding primer pair b), together with phosphate pool, the total lipid extracts were subjected to normal-phase liquid genomic DNA from cells of the parent strain (left) or from cells that chromatography (LC)/mass spectrometry (MS) analysis without prefraction- had replaced the HVO_1517 gene with the trpA sequence (middle) ation. as a template. (Right) PCR amplification was performed by using primers directed against the HVO_1517 coding region (i.e., primer LC/MS. Normal-phase LC-electron spray ionization (ESI)/MS of lipids was pair c), together with genomic DNA from cells of the parent strain performed by using an Agilent 1200 quaternary LC system coupled to a (parent) or the HVO_1517-deleted strain (⌬HVO_1517). Schematic Q-Star XL quadrupole time-of-flight (TOF) tandem mass spectrometer (Ap- diagrams showing the relative positions of the forward and reverse plied Biosystems, Foster City, CA). An Ascentis Si high-performance liquid ␮ primers in each primer pair appear below the panels. Note that chromatography (HPLC) column (5 m; 25 cm by 2.1 mm) was used. Mobile primer pairs a and b share the same reverse primer, while primer phase A consisted of chloroform-methanol-aqueous ammonium hydroxide pairs a and c share the same forward primer. (B) RT-PCR was (800:195:5, vol/vol/vol). Mobile phase B consisted of chloroform-methanol- performed by using primers directed at HVO_1517 (primer pair c) water-aqueous ammonium hydroxide (600:340:50:5, vol/vol/vol/vol). Mobile and genomic DNA from parent strain cells or cDNA or RNA from phase C consisted of chloroform-methanol-water-aqueous ammonium hy- HVO_1517-deleted cells as a template. (C) Deletion of H. volcanii droxide (450:450:95:5, vol/vol/vol/vol). The elution program consisted of the HVO_1517 affects the apparent molecular weight of the S-layer following: 100% mobile phase A was held isocratically for 2 min and then glycoprotein. Equivalent aliquots of H. volcanii cells lacking linearly increased to 100% mobile phase B over 14 min and held at 100% B HVO_1517 (⌬HVO_1517), cells of the parent strain (parent), or for 11 min. The LC gradient was then changed to 100% mobile phase C over cells of the same strain lacking aglD (⌬aglD) were separated by 3 min, held at 100% mobile phase C for 3 min, and finally returned to 100% 7.5% SDS-PAGE and Coomassie stained. Only the gel region con- mobile phase A over 0.5 min and held at 100% mobile phase A for 5 min. The taining the S-layer glycoprotein is shown. total LC flow rate was 300 ␮l/min. The postcolumn splitter diverted ϳ10% of the LC flow to the ESI source of the Q-Star XL mass spectrometer, with MS settings as follows: ion spray of Ϫ4,500 V, curtain of 20 lb/in2, gas source 1 of RESULTS 20 lb/in2, declustering potential of Ϫ55 V, and focusing potential of Ϫ150 V. Nitrogen was used as the collision gas for tandem MS (MS/MS) experiments. Deletion of H. volcanii HVO_1517 affects the S-layer glyco- Data acquisition and analysis were performed by using the instrument’s protein. Analyst QS software. To begin assessing the involvement of H. volcanii Mass spectrometry analyses of the S-layer glycoprotein from cells of the H. HVO_1517 in N glycosylation, the gene was deleted from the volcanii parent strain and of H. volcanii ⌬HVO_1517 cells were performed as genome as previously described (1, 4). In this so-called “pop- described elsewhere previously (3). in/pop-out” approach, the genomic copy of the gene of interest Proteolytic digestion of the H. volcanii S layer. S-layer resistance to the pro- in a strain auxotrophic for tryptophan is replaced by the H. ⌬ teolysis of cells of the H. volcanii parent strain and of H. volcanii HVO_1517 volcanii tryptophan synthase-encoding trpA sequence. The suc- cells (1 ml) grown to the mid-exponential phase was tested upon the addition of proteinase K (1 mg/ml) for up to 3 h. Aliquots (100 ␮l) were removed at 30-min cessful replacement of HVO_1517 by trpA was first verified at intervals during this window, and the proportion of intact S-layer glycoprotein the DNA level by PCR, using genomic DNA from the parent remaining was verified by densitometry following SDS-PAGE and Coomassie or the deletion strain as a template together with primer pair staining at each time point. Densitometry was performed by using EZQuant-Gel a (comprising a forward primer directed at a region within v.2.1 software. Values are expressed as percentages of the band intensity mea- HVO_1517 and a reverse primer directed at a downstream sured prior to protease treatment. Nucleotide sequence accession number. The sequence of H. volcanii AglJ has region) (Fig. 1A, left and middle) or primer pair b (comprising been deposited into the EMBL/GenBank/DDBJ database and assigned acces- the same reverse primer together with a forward primer di- sion number CAR66203. rected at a region within trpA) (Fig. 1A, left and middle). 5574 KAMINSKI ET AL. J. BACTERIOL.

While the parent strain contains the HVO_1517 sequence (Fig. metric analysis, provided support for HVO_1517 as adding the 1A, left), the gene was not detected in the deletion strain, first hexose of the N-linked pentasaccharide decorating the having been replaced by the trpA sequence (Fig. 1A, middle). S-layer glycoprotein. Although a detailed analysis of glycan- The absence of the HVO_1517 gene in the deletion strain was charged dolichol phosphates in H. volcanii will be addressed in further indicated by the failure to obtain a PCR product using a forthcoming study (Z. Guan et al., submitted for publica- primers directed against the coding region of this sequence tion), it is shown here that the H. volcanii parent strain pre- (Fig. 1A, right). sents a mass spectrometry profile that includes ion peaks of m/z The deletion of HVO_1517 was next confirmed at the RNA 1,011.744 {corresponding to the [M Ϫ H]Ϫ ion of hexose- level by RT-PCR performed using primers directed to the (C55)dolichol phosphate} and m/z 1,079.799 {corresponding to Ϫ Ϫ HVO_1517 coding region. In these reactions, cDNA derived the [M H] ion of hexose-(C60)dolichol phosphate; the from the parent or the deletion strain served as a template. No calculated mass of the [M Ϫ H]Ϫ ion is 1,079.805} (Fig. 3A, HVO_1517-derived band could be detected using cDNA from top). In addition, a major peak at m/z 1,055.7, corresponding to the HVO_1517 deletion strain (Fig. 1B, middle). In control a previously described sulfoglycolipid (22), was also observed. experiments, HVO_1517 was amplified by using genomic DNA When the same profile was considered for cells lacking from the parent strain as a template but not when RNA served HVO_1517 (Fig. 3A, bottom), the peaks corresponding to hex- as the template (Fig. 1B, left and right lanes, respectively). ose-modified (C55/C60)dolichol phosphate were also detected, To determine whether HVO_1517 is involved in protein although at barely detectable levels and approximately 6-fold glycosylation, the S-layer glycoprotein, a well-studied reporter less than what was observed for the parent strain (Fig. 3A, of H. volcanii N glycosylation (1–3, 17, 18, 23, 29, 30), was compare top and bottom insets). Indeed, the decrease in the characterized for cells lacking HVO_1517. Previous efforts (1, amount of hexose-modified (C55/C60)dolichol phosphate re- 29, 30) revealed that the deletion of agl genes encoding com- sulting from the deletion of HVO_1517 is also apparent when ponents of the H. volcanii N-glycosylation pathway often re- the relative heights of these peaks and that of the sulfoglyco- sulted in versions of the S-layer glycoprotein that migrate lipid are compared for each strain. In Fig. 3B, the MS/MS faster on SDS-PAGE gels than does the protein from parent spectrum confirming the structure of hexose-(C60)dolichol strain cells, due to absent or perturbed N glycosylation in the phosphate is shown. mutant cells. The migration of the S-layer glycoprotein in cells Given its role in N glycosylation, as demonstrated at the deleted of HVO_1517 was enhanced relative to the SDS- levels of both the dolichol carrier and the modified protein PAGE migration of the S-layer glycoprotein from cells of the target, HVO_1517 was renamed aglJ (GenBank accession num- parent strain yet was comparable to that seen in cells lacking ber CAR66203), in accordance with the nomenclature pro- AglD (Fig. 1C). AglD was shown previously to be involved in posed previously by Chaban et al. (7). the addition of the fifth sugar of the pentasaccharide N linked AglJ is involved in modifying only one of three H. volcanii to the S-layer glycoprotein (2). monosaccharide-charged dolichol phosphates. To better un- Deletion of HVO_1517 affects the composition of the glycan derstand the origin of the minor monosaccharide-modified N linked to the S-layer glycoprotein. To determine whether peaks associated with both the S-layer glycoprotein-derived HVO_1517 plays a role in N glycosylation, the S-layer glycopeptide and the dolichol phosphate carrier in the absence of protein from the HVO_1517 deletion mutant was examined by AglJ, the monosaccharide-charged (C60)dolichol phosphate mass spectrometry. In cells of the parent strain, at least two peak was isolated from both the parent and deletion strains, S-layer glycoprotein sequons (i.e., Asn-X-Ser/Thr motifs, and each one was examined in a single LC/MS run, over a where X is any residue but Pro) are modified by a pentasac- 40-min span. As reflected in Fig. 4, the single m/z 1,079 ion charide consisting of two hexose residues, two hexuronic acid peak from each strain could be resolved into three distinct residues, and a methyl ester of a hexuronic acid. As previously species, distinguished by their slightly different LC retention shown (2, 3, 17, 29, 30), mass spectrometry analysis of an times. When the relative levels of each peak from the parent Asn-13-containing S-layer glycoprotein-derived tryptic peptide strain were compared with their counterparts in the deletion isolated from cells of the parent strain revealed the presence of strain, it was noted that although the first and third fractions the peptide modified by the pentasaccharide (m/z 2,448.04) were not affected by the absence of AglJ, the second peak was (Fig. 2, top). In addition, other peptide peaks modified by reduced from a value of over 19,000 intensity units to 2,600 monosaccharides (m/z 1,743.95), disaccharides (m/z 1,919.86), intensity units in cells lacking AglJ, reflecting a 7-fold decrease trisaccharides (m/z 2,095.91), and tetrasaccharides (m/z in intensity (Fig. 4). This drop is comparable to the decrease in 2,285.97) were seen. In contrast, an examination of the Asn- the intensity of both the monosaccharide-modified S-layer gly- 13-linked glycan present on the same peptide derived from coprotein-derived peptide peak and the total monosaccharide- cells lacking HVO_1517 (Fig. 2, bottom) revealed the presence charged dolichol phosphate peak observed upon the deletion of only a minor peak corresponding to the monosaccharide- of aglJ (Fig. 2 and 3, respectively). modified peptide (m/z 1,743.83) as well as a major peak cor- One other H. volcanii Dpm1 homologue generates monosac- responding to the nonmodified peptide (m/z 1,581.77). Peaks charide-modified dolichol phosphate. As H. volcanii was orig- corresponding to di-, tri-, tetra-, and pentasaccharide-modified inally reported to contain five Dpm1 homologues (1), the pu- peptides were completely absent in this sample. tative ability of these sequences to generate monosaccharide- Deletion of HVO_1517 is manifested at the monosaccharide- charged dolichol phosphate, including that minor species charged dolichol phosphate level. The isolation of the dolichol observed in the absence of AglJ, was next considered. Accord- phosphate pool from the H. volcanii parent strain and from the ingly, the monosaccharide-charged (C60)dolichol phosphate same cells deleted of HVO_1517, followed by mass spectro- peak pools from H. volcanii strains lacking either HVO_2601, VOL. 192, 2010 HALOFERAX VOLCANII AglJ 5575

FIG. 2. Matrix-assisted laser desorption ionization (MALDI)–TOF analysis of an Asn-13-containing H. volcanii S-layer glycoprotein-derived glycopeptide. The MALDI-TOF spectra of the Asn-13-containing tryptic peptides derived from the S-layer glycoprotein from cells of the parent strain (top) or of the HVO_1517-deleted strain (⌬HVO_1517) (bottom) are shown. The components of the glycopeptide-associated sugar residues, as well as the glycopeptide amino acid sequence, are shown in the inset box, while the glycan moieties decorating the peptide peaks are marked on the MALDI-TOF spectra accordingly.

first identified as Dpm1-A, or HVO_1613, first identified as to add a monosaccharide to dolichol phosphate although not Dpm1-D, were each examined in a single LC/MS run over a that subunit onto which additional sugars are added. 40-min span. Whereas the deletion of HVO_2601 had no effect In cells lacking AglJ, S-layer integrity is compromised. To on the monosaccharide-charged (C60)dolichol phosphate pro- assess whether the observed modification of the S-layer glyco- file (not shown), cells lacking HVO_1613 did not generate the protein glycan in AglJ-lacking cells also affected the integrity of small monosaccharide-charged dolichol phosphate species re- the S layer surrounding H. volcanii cells (thought to be com- tained at the 16.06-min point in the parent strain (Fig. 5). The posed solely of the S-layer glycoprotein [23]), parent strain remaining two Dpm1 homologues were not considered in these cells and cells of the same strain deleted of aglJ were chal- experiments, since Dpm1-B was previously reannotated as lenged with proteinase K for up to 3 h. The level of intact AglE, shown to be responsible for adding pentasaccharide S-layer glycoprotein remaining at various time points within subunit 4 (3), while the final Dpm1 homologue, HVO_A0194, this window was revealed by densitometry following SDS- was originally shown to be transcribed only under heat shock PAGE and Coomassie staining. The results show that substan- conditions but not during log-phase growth in complete me- tially more full-length S-layer glycoprotein from the parent dium (1). As such, HVO_A0194 could not contribute to the strain survived the proteinase K challenge than from cells monosaccharide charging of dolichol phosphate processing ob- lacking AglJ (Fig. 6). Indeed, densitometric quantitation of served here. Hence, it can be concluded that HVO_1613 serves repeats of the experiment revealed that whereas almost 51% of 5576 KAMINSKI ET AL. J. BACTERIOL.

FIG. 3. The absence of HVO_1517 affects the level of monosaccharide-modified dolichol phosphate. (A) Normal-phase LC-ESI/MS analysis in the negative-ion mode of the total lipid extracts from cells of the parent strain (parent) (top) and the HVO_1517-deleted strain (⌬HVO_1517) (bottom). amu, atomic mass units. The mass spectra were averaged from the spectra obtained between the 15.5- and 17.5-min retention times. The inset in each panel represents a 5-fold expansion of the hexose-(C60)dolichol phosphate region of the profile. (B) MS/MS verification of dolichylphosphate-hexose by collision-induced dissociation of its [M Ϫ H]Ϫ ion at m/z 1,079.9. the starting S-layer glycoprotein in parent strain cells survived the S-layer glycoprotein in the absence of AglJ compromises 90 min of incubation with the protease, only 13% of the same the proper assembly of the protein shell surrounding H. volca- protein remained after this interval in AglJ-lacking cells. It nii cells, as was previously observed in the case of cells lacking thus appears that the reduced N glycosylation experienced by AglF, AglG, AglI, or AglM (29, 30). VOL. 192, 2010 HALOFERAX VOLCANII AglJ 5577

FIG. 4. Only one of three monosaccharide-modified dolichol phosphates is affected by a lack of AglJ. Normal-phase LC-extracted ion chromatograms (EIC) of the dolichylphosphate-hexose [M Ϫ H]Ϫ ion at m/z 1,079.8 from the parent strain (top) and the ⌬aglJ strain (bottom) are shown. The peaks at different retention times suggest the existence of three different dolichylphosphate-hexose species. Note the different scales used on the ordinates of the two graphs, highlighting that the 16.2-min peak is reduced 7.3-fold in the mutant compared with the parent strain, suggesting that AglJ is specific for the formation of this particular monosaccharide-modified dolichol phosphate species.

DISCUSSION charged dolichol phosphate species. The same is likely true for the minor peak observed for the deletion strain at the position In the present study, mass spectrometry analysis at both the normally occupied by the AglJ-processed peak, which could glycopeptide and dolichol phosphate carrier levels assigned appear due to the actions of another glycosyltransferase inef- AglJ a role in adding the first, as-yet-unidentified hexose to the ficiently filling the void left in the absence of AglJ. Alterna- pentasaccharide decorating the H. volcanii S-layer glycopro- tively, this minor monosaccharide-modified dolichol phosphate tein. However, in contrast to previous studies addressing the may be naturally present but is unmasked only now, in the other predicted glycosyltransferases participating in H. volcanii absence of an AglJ-catalyzed glycosylation of dolichol phos- N glycosylation (i.e., AglD, AglE, AglG, and AglI), where the phate, meaning that four different monosaccharide-modified deletion of the encoding gene led to the appearance of N- dolichol phosphate pools would exist in H. volcanii, namely, linked glycans totally lacking the sugar subunit added by the those generated by the glycosyltransferases AglJ, HVO_1613, glycosyltransferase in question (2, 3, 29), small amounts of hexose-modified dolichol phosphate and S-layer glycoprotein- and two others, which are currently unidentified. The results derived peptide were observed for cells lacking AglJ. confirm that HV_2601, previously identified as a Dpm1 homo- This observation can be explained upon a detailed analysis logue (1), does not serve such a role. It is also conceivable that of the monosaccharide-charged dolichol phosphate pool in the AglJ does not act alone in adding the first hexose of the parent and aglJ deletion strains. Such an analysis revealed the N-linked pentasaccharide such that the relevant monosaccha- existence of three distinct monosaccharide-modified lipid car- ride-modified lipid carrier and protein target observed for cells riers, reminiscent of a previous study reporting that radio- lacking AglJ reflect the residual contribution of a second pro- chemical amounts of glucose- and galactose-charged dolichol tein involved in adding this first pentasaccharide sugar subunit. phosphate could be detected in H. volcanii (14). Of the three Regardless of the agent(s) responsible for generating the monosaccharide-modified lipid carriers identified in the monosaccharide-charged dolichol phosphate species observed present study, the deletion of AglJ affected only one species. for the aglJ deletion strain, no additional saccharide subunits The other two minor peaks unaffected by the absence of AglJ are added to that minor monosaccharide-modified dolichol would thus apparently be the products of different glycosyl- phosphate carrier seen at the position of the AglJ-processed transferases. Indeed, it was shown that HVO_1613 is respon- peak in this mutant (or, for that matter, to any of the other sible for generating one of these minor monosaccharide- minor populations of monosaccharide-modified carriers). 5578 KAMINSKI ET AL. J. BACTERIOL.

FIG. 5. HVO_1613, a Dpm1 homologue, is responsible for generating one of three H. volcanii monosaccharide-modiﬁed dolichol phosphates. Normal-phase LC EICs of the dolichylphosphate-hexose [M Ϫ H]Ϫ ion at m/z 1,079.8 from the parent strain (top) and from ⌬aglJ (middle) and ⌬HVO_1613 cells (bottom) are shown. In the absence of HVO_1613, the ﬁrst dolichylphosphate-hexose species (16.06 min) is absent.

Moreover, no additional saccharide subunits are bound to the AglJ-processed, monosaccharide-charged dolichol phosphate monosaccharide N linked to the S-layer glycoprotein in the aglJ carrier. Still, the possibility remains that the hexose found at deletion strain. Thus, it seems unlikely that the monosaccha- pentasaccharide position 5 is derived from one of the non- ride added to the lipid carrier found at the same position as the AglJ-dependent monosaccharide-charged dolichol phos- AglJ-processed lipid carrier in the deletion strain represents an phates. AglD, previously shown to be involved in adding the alternative linking sugar of a pentasaccharide variant decorat- final pentasaccharide subunit (2), could fulfill such a role. Fi- ing the S-layer glycoprotein. Likewise, the other minor nally, the minor amount of monosaccharide-modified dolichol monosaccharide-modified dolichol phosphates do not appear phosphate carriers generated through the actions of enzymes to participate in generating the pentasaccharide N linked to other than AglJ could participate in the biosynthesis of other the S-layer glycoprotein. Moreover, given that sugar subunits 2 glycoconjugates, such as glycolipids. Indeed, the biogenesis of and 3 of the S-layer-linked pentasaccharide are hexuronic acids H. volcanii glycolipids, a process of which little is known, was (2) while subunit 4 is a methyl ester of hexuronic acid (17), it previously shown not to involve any of the Agl glycosyltrans- is also unlikely that the hexose-charged dolichol phosphates ferases (20). generated through the actions of glycosyltransferases other Previous results have shown that in cells deleted of aglG, than AglJ serve as lipid carriers for those sugars ultimately encoding the predicted glycosyltransferase responsible for add- found at pentasaccharide position 2, 3, or 4. As such, it appears ing the second sugar of the pentasaccharide N linked to the H. that the two hexuronic acid and the methyl ester of hexuronic volcanii S-layer glycoprotein, significant amounts of monosac- acid components of the N-linked pentasaccharide are derived charide-modified protein were generated (29). As such, the from soluble activated species sequentially added onto an as-yet-unidentified archaeal flippase(s) responsible for delivering the glycan-charged dolichol phosphate carrier across the plasma membrane as well as AglB, the archaeal oligosaccharyltransferase responsible for transferring the “flipped” dolichol phosphate-bound glycan to the protein target, are apparently able to recognize even the AglJ-processed, monosaccharide- modified substrate. Likewise, in the present study, the minor amounts of the monosaccharide-modified S-layer glycoprotein observed for the ⌬aglJ strain point to the abilities of the H. volcanii flippase(s) and AglB to deliver even smaller amounts FIG. 6. The S layer surrounding H. volcanii cells is protease sensi- of non-AglJ-processed monosaccharides from their dolichol tive in cells deleted of aglJ. Shown are data for cells of the parent strain phosphate carriers to target proteins. Moreover, the finding (parent) (top) or cells of the same strain lacking AglJ (⌬aglJ) or that that the monosaccharide-modified S-layer glycoprotein was were challenged with 1 mg/ml proteinase K (pK) at 37°C (bottom). ⌬aglJ Aliquots were removed immediately prior to incubation with the pro- observed for the strain, likely bearing an alternative tease and at subsequent intervals up to 3 h and examined by 7.5% sugar from that found on the monosaccharide-modified S-layer SDS-PAGE. glycoprotein population seen in native cells, is indicative of the VOL. 192, 2010 HALOFERAX VOLCANII AglJ 5579 relaxed specificity of H. volcanii AglB for the linking sugar, and P. Castric. 2002. Glycosylation of Pseudomonas aeruginosa 1244 pilin: similar to what was seen previously with bacterial oligosac- glycan substrate specificity. Mol. Microbiol. 46:519–530. 10. Faridmoayer, A., M. A. Fentabil, D. C. Mills, J. S. Klassen, and M. F. charyltransferases (9–11). Indeed, such a relaxed specificity of Feldman. 2007. Functional characterization of bacterial oligosaccharyltrans- the archaeal oligosaccharyltransferase was previously sug- ferases involved in O-linked protein glycosylation. J. Bacteriol. 189:8088– 8098. gested in light of reports of the Halobacterium salinarum S- 11. Faridmoayer, A., M. A. Fentabil, M. F. Haurat, W. Yi, R. Woodward, P. G. layer glycoprotein being modified by two different N-linked Wang, and M. F. Feldman. 2008. Extreme substrate promiscuity of the glycans, each bearing a unique linking sugar, despite the seem- Neisseria oligosaccharyltransferase involved in protein O-glycosylation. J. Biol. Chem. 283:34596–34604. ing presence of a single AglB protein in this species (15, 16). 12. Helenius, A., and M. Aebi. 2004. Roles of N-linked glycans in the endoplas- In conclusion, AglJ can now be added to the growing list of mic reticulum. Annu. Rev. Biochem. 73:1019–1049. H. volcanii components shown to contribute to protein N gly- 13. Igura, M., N. Maita, J. Kamishikiryo, M. Yamada, T. Obita, K. Maenaka, and D. Kohda. 2008. Structure-guided identification of a new catalytic motif cosylation in this haloarchaeon that includes AglB, AglD, of oligosaccharyltransferase. EMBO J. 27:234–243. AglE, AglF, AglG, AglI, AglM, and AglP (2, 3, 12, 29, 30), all 14. Kuntz, C., J. Sonnenbichler, I. Sonnenbichler, M. Sumper, and R. Zeitler. 1997. Isolation and characterization of dolichol-linked oligosaccharides from of which (with the exception of AglD) are encoded by a single Haloferax volcanii. Glycobiology 7:897–904. agl gene cluster (27). These findings, along with similar efforts 15. Lechner, J., and F. Wieland. 1989. Structure and biosynthesis of prokaryotic addressing other members of the Archaea, such as Methano- glycoproteins. Annu. Rev. Biochem. 58:173–194. 16. Magidovich, H., and J. Eichler. 2009. Glycosyltransferases and oligosac- coccus voltae (7, 8, 21), Methanococcus maripaludis (25), and charyltransferases in Archaea: putative components of the N-glycosylation Pyrococcus furiosus (13), are contributing to a better under- pathway in the third domain of life. FEMS Microbiol. Lett. 300:122–130. standing of archaeal N glycosylation. 17. Magidovich, H., S. Yurist-Doutsch, Z. Konrad, V. V. Ventura, A. Dell, P. G. Hitchen, and J. Eichler. 2010. AglP is a S-adenosyl-L-methionine-dependent methyltransferase that participates in the N-glycosylation pathway of ACKNOWLEDGMENTS Haloferax volcanii. Mol. Microbiol. 76:190–199. 18. Mengele, R., and M. Sumper. 1992. Drastic differences in glycosylation of Plasmid pTA131 and H. volcanii WR536 (H53) cells were kindly related S-layer glycoproteins from moderate and extreme halophiles. J. Biol. provided by Moshe Mevarech (Tel Aviv University). J.E. is supported Chem. 267:8182–8185. by the Israel Science Foundation (grant 30/07) and the U.S. Army 19. Mevarech, M., and R. Werczberger. 1985. Genetic transfer in Halobacterium volcanii. J. Bacteriol. 162:461–462. Research Office (grant W911NF-07-1-0260). 20. Naparstek, S., E. Vinagradov, and J. Eichler. 2010. Different glycosyltrans- The mass spectrometry facility in the Department of Biochemistry of ferases are involved in lipid glycosylation and protein N-glycosylation in the the Duke University Medical Center and Z.G. are supported by Lipid halophilic archaeon Haloferax volcanii. Arch. Microbiol. 192:581–584. Maps large-scale collaborative grant GM-069338 from the NIH. A.D. 21. Shams-Eldin, H., B. Chaban, S. Niehus, R. T. Schwarz, and K. F. Jarrell. is supported by the Biotechnology and Biological Sciences Research 2008. Identification of the archaeal alg7 gene homolog (N-acetylglu- Council (grants BBF0083091 and BBC5196701). L.K. is the recipient cosamine-1-phosphate transferase) of the N-linked glycosylation system by of a Negev-Zin Associates scholarship. V.V.V. is supported by Dstl. cross-domain complementation in yeast. J. Bacteriol. 190:2217–2220. 22. Sprott, G. D., S. Larocque, N. Cadotte, C. J. Dicaire, M. McGee, and J. R. Brisson. 2003. Novel polar lipids of halophilic eubacterium Planococcus H8 REFERENCES and archaeon Haloferax volcanii. Biochim. Biophys. Acta 1633:179–188. 1. Abu-Qarn, M., and J. Eichler. 2006. Protein N-glycosylation in Archaea: 23. Sumper, M., E. Berg, R. Mengele, and I. Strobel. 1990. Primary structure and defining Haloferax volcanii genes involved in S-layer glycoprotein glycosyla- glycosylation of the S-layer protein of Haloferax volcanii. J. Bacteriol. 172: tion. Mol. Microbiol. 61:511–525. 7111–7118. 2. Abu-Qarn, M., S. Yurist-Doutsch, A. Giordano, A. Trauner, H. R. Morris, P. 24. Szymanski, C. M., and B. W. Wren. 2005. Protein glycosylation in bacterial Hitchen, O. Medalia, A. Dell, and J. Eichler. 2007. Haloferax volcanii AglB mucosal pathogens. Nat. Rev. Microbiol. 3:225–237. and AglD are involved in N-glycosylation of the S-layer glycoprotein and 25. VanDyke, D. J., J. Wu, S. M. Logan, J. F. Kelly, S. Mizuno, S. I. Aizawa, and proper assembly of the surface layer. J. Mol. Biol. 374:1224–1236. K. F. Jarrell. 2009. Identification of genes involved in the assembly and 3. Abu-Qarn, M., A. Giordano, F. Battaglia, A. Trauner, P. Hitchen, H. R. attachment of a novel flagellin N-linked tetrasaccharide important for mo- Morris, A. Dell, and J. Eichler. 2008. Identification of AglE, a second tility in the archaeon Methanococcus maripaludis. Mol. Microbiol. 72:633– glycosyltransferase involved in N-glycosylation of the Haloferax volcanii S- 644. layer glycoprotein. J. Bacteriol. 190:3140–3146. 26. Weerepana, E., and B. Imperiali. 2006. Asparagine-linked protein glycosyl- 4. Allers, T., H. P. Ngo, M. Mevarech, and R. G. Lloyd. 2004. Development of ation: from eukaryotic to prokaryotic systems. Glycobiology 16:91R–101R. additional selectable markers for the halophilic archaeon Haloferax volcanii 27. Yurist-Doutsch, S., and J. Eichler. 2009. Manual annotation, transcriptional based on the leuB and trpA genes. Appl. Environ. Microbiol. 70:943–953. analysis and protein expression studies reveal novel genes in the agl cluster 5. Burda, P., and M. Aebi. 1999. The dolichol pathway of N-linked glycosyla- responsible for N-glycosylation in the halophilic archaeon Haloferax volcanii. tion. Biochim. Biophys. Acta 1426:239–257. J. Bacteriol. 191:3068–3075. 6. Calo, D., L. Kaminski, and J. Eichler. 2010. Protein glycosylation in Ar- 28. Yurist-Doutsch, S., B. Chaban, D. VanDyke, K. F. Jarrell, and J. Eichler. chaea: sweet and extreme. Glycobiology 20:1065–1076. 2008. Sweet to the extreme: protein glycosylation in Archaea. Mol. Micro- 7. Chaban, B., S. Voisin, J. Kelly, S. M. Logan, and K. F. Jarrell. 2006. biol. 68:1079–1084. Identification of genes involved in the biosynthesis and attachment of Meth- 29. Yurist-Doutsch, S., M. Abu-Qarn, F. Battaglia, H. R. Morris, P. G. Hitchen, anococcus voltae N-linked glycans: insight into N-linked glycosylation path- A. Dell, and J. Eichler. 2008. aglF, aglG and aglI, novel members of a gene ways in Archaea. Mol. Microbiol. 61:259–268. cluster involved in the N-glycosylation of the Haloferax volcanii S-layer gly- 8. Chaban, B., S. M. Logan, J. F. Kelly, and K. F. Jarrell. 2009. AglC and AglK coprotein. Mol. Microbiol. 69:1234–1245. are involved in biosynthesis and attachment of diacetylated glucuronic acid 30. Yurist-Doutsch, S., H. Magidovich, V. V. Ventura, P. G. Hitchen, A. Dell, and to the N-glycan in Methanococcus voltae. J. Bacteriol. 191:187–195. J. Eichler. 2010. N-glycosylation in Archaea: on the coordinated actions of 9. DiGiandomenico, A., M. J. Matewish, A. Bisaillon, J. R. Stehle, J. S. Lam, Haloferax volcanii AglF and AglM. Mol. Microbiol. 74:1047–1058.

Chapter 3.3

"Identification of residues important for the activity of Haloferax

volcanii AglD, a component of the archaeal

N-glycosylation pathway"

Lina Kaminski and Jerry Eichler Archaea. 2010. 6;2010:315108

Abstract:

In Haloferax volcanii, AglD adds the final hexose to the N-linked pentasaccharide decorating the S-layer glycoprotein. Not knowing the natural substrate of the glycosyltransferase, together with the challenge of designing assays compatible with hypersalinity, has frustrated efforts at biochemical characterization of

AglD activity. To circumvent these obstacles, an in vivo assay designed to identify amino acid residues important for AglD activity is described. In the assay, restoration of AglD function in a Hfx. volcanii aglD deletion strain transformed to express plasmid-encoded versions of AglD, generated through site-directed mutagenesis at positions encoding residues conserved in archaeal homologues of AglD, is reflected in the behavior of a readily detectable reporter of N-glycosylation. As such and Asp110,

Asp112 were designated as elements of the DXD motif of AglD, a motif that interacts with metal cations associated with nucleotide-activated sugar donors, while Asp201 was predicted to be the catalytic base of the enzyme.

Hindawi Publishing Corporation Archaea Volume 2010, Article ID 315108, 9 pages doi:10.1155/2010/315108

Research Article Identiﬁcation of Residues Important for the Activity of Haloferax volcanii AglD, a Component of the Archaeal N-Glycosylation Pathway

Lina Kaminski and Jerry Eichler

Department of Life Sciences, Ben Gurion University, P.O. Box 653, Beersheva 84105, Israel

Correspondence should be addressed to Jerry Eichler, [email protected]

Received 11 January 2010; Accepted 10 February 2010

Academic Editor: Julie Maupin-Furlow

Copyright © 2010 L. Kaminski and J. Eichler. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

In Haloferax volcanii, AglD adds the final hexose to the N-linked pentasaccharide decorating the S-layer glycoprotein. Not knowing the natural substrate of the glycosyltransferase, together with the challenge of designing assays compatible with hypersalinity, has frustrated efforts at biochemical characterization of AglD activity. To circumvent these obstacles, an in vivo assay designed to identify amino acid residues important for AglD activity is described. In the assay, restoration of AglD function in an Hfx. volcanii aglD deletion strain transformed to express plasmid-encoded versions of AglD, generated through site-directed mutagenesis at positions encoding residues conserved in archaeal homologues of AglD, is reflected in the behavior of a readily detectable reporter of N-glycosylation. As such Asp110 and Asp112 were designated as elements of the DXD motif of AglD, a motif that interacts with metal cations associated with nucleotide-activated sugar donors, while Asp201 was predicted to be the catalytic base of the enzyme.

1. Introduction functions fulfilled by family members [9, 10]. Furthermore, the different GT families can be clustered based on whether Although the presence of N-glycosylated proteins in Archaea the canonical GT-A or GT-B fold is employed and whether has been known for over 30 years [1], the pathways responsi- sugar stereochemistry is retained or inverted upon addition ble for this posttranslational modification have only recently of a glycosyl donor [11]. Still, the ability to predict the been addressed. In Methanococcus voltae, Methanococcus function of a given GT or to define its catalytic mechanism maripaludis, and Haloferax volcanii, products of the agl remains a challenge. This is particularly true in the case genes have been shown to participate in the assembly of of the GT2 family, an ancient group of GT-A fold-bearing oligosaccharides decorating various glycoproteins in these GTs containing over 10,000 members derived from various species [2–4]. At present, however, apart from the oligosac- sources and serving at least 12 distinct functions [11, 12]. charyltransferase, AglB [5–7], virtually nothing is known Like all GT-A fold-bearing GTs, GT2 family members con- of the catalytic workings of the different Agl proteins. Of tain a DXD signature motif, shown to interact with a divalent the Hfx. volcanii Agl proteins identified to date, at least cation (usually Mg2+ or Mn2+) that facilitates the leaving of five (i.e., AglD, AglE, AglG, AglI, and AglJ) are predicted the nucleoside diphosphate group of a nucleotide-activated to act as glycosyltransferases (GTs), enzymes that catalyze sugar donor as part of the SN2-like displacement mechanism the formation of glycosidic bonds through the transfer of believed to be employed by these enzymes [11, 13–17]. The the sugar moieties from nucleotide-activated saccharides to DXD motif also serves to divide the GT-A fold into two appropriate targets [8]. portions. The N-terminal portion, containing the sequence Based on their amino acid similarities, GTs can be that assigns the protein to the GT2 family [18], binds the classified into 91 family groups (http://www.cazy.org/fam/ nucleotide-activated sugar donor [19–22]. By contrast, the acc GT.html; January, 2009), varying in size and number of C-terminal portion is highly variable and generally serves 2 Archaea to recognize the acceptor [17]. Despite such variability, was confirmed by sequencing, performed both before and the C-terminal portion of GT2 family members includes following introduction of plasmid-encoded mutated aglD a conserved Asp or Glu residue that presumably serves as into Hfx. volcanii. the catalytic base, thought to assist in the protonation of the nucleophilic hydroxyl group of the acceptor saccharide 2.4. Other Methods. Periodic acid-Schiff (PAS) reagent glyco- [11, 21, 23–25]. protein staining was performed as described previously [30]. The GT2 glycosyltransferase family includes Hfx. volcanii Immunoblots were performed using polyclonal antibodies AglD, previously shown to participate in adding the final raised against the C. thermocellum CBD (obtained from Ed hexose to the pentasaccharide comprising two hexoses, two Bayer, Weizmann Institute of Science; 1 : 10,000). Antibody hexuronic acids, and a methylated ester of hexuronic acid binding was detected using goat antirabbit horseradish decorating at least two sequons of the S-layer glycoprotein [26, 27]. However, due to the fact that its natural substrates peroxidase-(HRP-) conjugated antibodies (1 : 4000, BioRad, have yet to be defined and the challenge of devising in vitro Hercules, CA) and an ECL-enhanced chemiluminescence kit assays for haloarchaeal enzymes due to their hypersaline (Amersham, Buckingham, UK). requirements, little is known of the catalytic workings of AglD. Towards remedying the situation, an in vivo approach 3. Results has been developed in which the ability of plasmid-encoded versions of AglD, modified through site-directed mutagene- 3.1. AglD Activity in AglD-Deleted Hfx. volcanii Cells Is sis, to restore the absent function to an aglD deletion strain, Restored Upon Complementation with Plasmid-Encoded AglD. was tested. Results obtained employing this novel assay point As a first step towards describing AglD function, efforts were to Asp110-Thr111-Asp112 as corresponding to the DXD directed at creating an assay to allow for characterization of motif and Asp201 as corresponding to the catalytic base of the activity of the enzyme. Accordingly, cells deleted of the Hfx. volcanii AglD. encoding gene were transformed with a plasmid encoding a version of the protein designed to include an N-terminally fused Clostridium thermocellum cellulose-binding domain 2. Methods (CBD) [29]. As previously reported [29], an 85 kDa band, 2.1. Strains and Growth Conditions. The Hfx. volcanii back- corresponding to the predicted molecular mass of the 17 kDa ground strain WR536 (H53) and the same strain deleted CBD moiety and the 68 kDa AglD protein, was expressed for aglD were grown in complete medium containing in the transformed cells and recognized in an immunoblot using antiCBD antibodies (not shown). 3.4MNaCl,0.15MMgSO4·7H20, 1 mM MnCl2,4mMKCl, Deletion of aglD results in the absence of the final hexose 3mMCaCl2, 0.3% (w/v) yeast extract, 0.5% (w/v) tryptone and 50 mM Tris-HCl, pH 7.2, at 42◦C[28]. A complete of the pentasaccharide decorating the Hfx. volcanii S-layer description of the aglD deletion strain and the protocol used glycoprotein [26]. As such, the S-layer glycoprotein in the to delete the gene have been previously published [5]. deletion strain migrates faster in SDS-PAGE than does the native protein in the background strain [5]. Moreover, the S-layer glycoprotein is not recognized by PAS glycostain in 2.2. In Vivo AglD Assay. To assay AglD activity, Hfx. volcanii the mutant strain. However, as reflected in Figure 1, the cells deleted of aglD [5]weretransformedtoexpress S-layer glycoprotein from cells of the aglD-deleted strain plasmid-encoded versions of AglD that included an N- transformed to express CBD-tagged AglD migrated to the terminally fused Clostridium thermocellum cellulose-binding same position as did the protein from the background strain domain (CBD) [29]. To introduce nonnative residues into and was similarly PAS-stained. As such, complementation AglD, the plasmid-encoded version of aglD (GenBank acces- of Hfx. volcanii cells lacking aglD with an AglD-encoding sion number CAM91696.1) was modified by site-directed plasmid restores the absent activity to the deletion strain. mutagenesis. Restoration of AglD function lost as a result To demonstrate the involvement of a given AglD residue of deletion of the genomic copy of the encoding gene in the activity of the enzyme, the return of AglD activity to was determined by the ability of the transformed cells to the aglD deletion strain, upon introduction of a plasmid- reverse the enhanced SDS-PAGE migration of the S-layer encoded version of AglD modified at the amino acid glycoprotein and loss of PAS glycostaining of the same position in question, was assessed. In these experiments, reporter, that is, novel traits of the S-layer glycoprotein that the SDS-PAGE migration of the S-layer glycoprotein from appeared in cells lacking AglD. cells of the background strain, from cells deleted of aglD, and from AglD-lacking cells transformed to express select 2.3. Site-Directed Mutagenesis. Mutated versions of aglD mutant AglD proteins was addressed. In addition, the S- were generated by site-directed mutagenesis using the layer glycoprotein in each of the three populations of Hfx. Quikchange (Stratagene) protocol, performed according volcanii cells was subjected to PAS glycostaining. While this in to the manufacturer’s instructions, with plasmid pWL- vivo approach cannot distinguish between residues necessary CBD-AglD, encoding CBD-AglD [29], serving as template. for catalytic activity from those important for proper AglD Oligonucleotide primers used to introduce the various muta- folding, it, nonetheless, offers a facile route for identifying tions are listed in Supplementary Table 1 (available online important AglD residues until such time as AglD activity can at doi:10.1155/2010/315108). The introduction of mutations be directly assayed in vitro. Archaea 3

ΔaglD/ express AglD D110A resulted in both the failure of plasmid- bkgnd ΔaglD CBD-aglD encoded AglD to restore S-layer glycoprotein migration to the position of this reporter in the background strain as CBB well as the lost ability of PAS glycostain to label the S- layer glycoprotein. The same was true in cells transformed to express AglD D110E, a mutation that retains the negative PAS charge at this position (Figure 3), or upon introduction of an Asn residue at this position (not shown). Figure 1: aglD-complemented Hfx. volcanii cells regain the ability When AglD Asp112 of the plasmid-encoded protein was to properly glycosylate the S-layer glycoprotein. The protein replaced with an Asn, no recovery of AglD function in contents of cells of the WR536 background strain (bkgnd), the the transformed aglD deletion strain was realized, reflected same strain deleted of aglD (ΔaglD) or the AglD-lacking strain in the inability of cells expressing the mutated version of transformed with a plasmid encoding CBD-AglD (ΔaglD/CBD- AglD to restore SDS-PAGE migration and PAS glycostaining aglD) were separated by 5% SDS-PAGE and the S-layer glycoprotein was detected by Coomassie stain (CBB) or periodic acid-Schiff of the S-layer glycoprotein, as realized in the background (PAS) reagent. In the presence of CBD-AglD, the migration and strain (Figure 3). By contrast, transformation of the deletion positive glycostaining of the S-layer glycoprotein are as observed in strain to express AglD D112E led to a restoration of SDS- the background strain. PAGE migration of the S-layer glycoprotein to the position seen in background cells but only a partial recovery (6% ± 0.5% (standard deviation), n = 3) of PAS glycostaining (Figure 3). 3.2. Identification of Conserved AglD Residues. To select The importance of Asp110 and Asp112 for Hfx. volcanii candidate residues for site-directed mutagenesis, the Hfx. AglD activity points to these two residues as comprising volcanii AglD sequence was aligned with selected homol- the DXD motif found in GT-A fold-bearing GTs. However, ogous archaeal sequences using ClustalW (http://www while the Asp residue at position 110 is apparently essential .ebi.ac.uk/Tools/clustalw2/index.html). It should be noted, for activity, the presence of Glu at position 112 yields however, that it is not yet known whether the various a functional enzyme that apparently acts differently from homologues considered indeed catalyze the same reaction as the native enzyme, as reflected in the limited PAS staining does Hfx. volcanii AglD. Indeed, it remains to be confirmed detected with this mutant. that N-glycosylation occurs in all of the species listed. Nonetheless, such alignment revealed the presence of a 3.4. Asp201 Is Likely the Catalytic Base of AglD. In addition stretch of amino acids in the N-terminal region of AglD to the DXD motif considered above, the activity of GT2 showing substantial overlap with similarly situated regions family members also relies on an Asp or Glu residue in the various archaeal homologues considered (Figure 2). found in the acceptor-binding domain of the protein. First In the Hfx. volcanii protein, this stretch corresponds to identified in the solved three-dimensional structure of the region between Asp110 and Glu203, a portion of the Bacillus subtilis SpsA as Asp191 [20], this residue and its protein previously localized to the cytoplasm [29]and equivalents in other GTs are thought to serve as the base which includes seven residues absolutely conserved in the catalyst in the direct displacement mechanism apparently sequences considered, namely, Asp110, Asp112, Asp133, employed by these enzymes [11, 20]. To identify the Hfx. Arg139, Arg152, Asp173, and Gly177. Between Asp133 and volcanii AglD equivalent of B. subtilis SpsA Asp191, the Arg139, between Trp198 and Glu203, and in the region sequence of the soluble region of AglD (residues 1–259) surrounding Asp173 and Gly177, several highly conserved was aligned with the sequence of B. subtilis SpsA, as well as residues were also detected. To determine whether any of with those of the other GT2 enzymes where the functional these residues contribute to AglD function, the correspond- equivalent of B. subtilis SpsA Asp191 is known, namely, ing aglD codons were modified by site-directed mutagenesis Sinorhizobium meliloti ExoM and Salmonella enterica WbbE. using the primer pairs listed in Supplementary Table 1, Earlier site-directed mutagenesis efforts had revealed ExoM and the ability of plasmid-encoded versions of the mutant Asp187 and WbbE Glu180 to serve the same role as SpsA proteins to restore AglD function in the aglD deletion strain Asp191 [21, 22]. Tcoffee (http://tcoffee.vital-it.ch/cgi-bin/ was considered. Tcoffee/ tcoffee cgi / index.cgi? stage1 =1anddaction=TCOF- FEE::Regular) aligned AglD Asp201 with SpsA Asp191, 3.3. The DXD Catalytic Motif of AglD Likely Comprises Asp110 ExoM Asp187 and WbbE Glu180. The MAFFT program and Asp 112. The GT-A fold found in GT2 family members (v6.531b; http://www.ebi.ac.uk/Tools/mafft/index.html) also includes a DXD motif that contributes to the catalytic activity aligned AglD Asp201 with the same SpsA, ExoM and of the enzyme [11, 13, 14, 16, 17]. Sequence alignment-based WbbE residues. On the other hand, alignment of archaeal examination of the Hfx. volcanii AglD sequence points to homologues of Hfx. volcanii AglD using the ClustalW Asp110-Thr111-Asp112 as comprising this motif (Figure 2). (Figure 2), Tcoffee or MAFFT programs revealed that To directly test this hypothesis, the site-directed mutagenesis AglD Asp173 but not Asp201 is conserved. In all cases, the approach described above was enlisted. Figure 3 addresses programs consulted were used with their default settings. the effects of replacing either AglD Asp110 or Asp112 As a next step towards identifying the AglD equivalent with other residues. Transformation of AglD-lacking cells to of SpsA Asp191, the importance of Asp173, Asp195, and 4 Archaea

Hvol 108 Y F DT DL A T DMR HL E E L V E R V R S G E Y DA A T G S R W- - MP DR V A DR P R K R G V P S R A Y NG L V R L F L R - - S DL R DHQC G F K A F S R E A F E A L R DD V - E D NHWF WD T E M 204 * * * * * * Aful 143 Y MDV DL A T DL S HL K E L V DA I I V E G Y DF S T G S R L - - MK E S QT DR P A K R E I A S R G Y NF L V R L F L G - - S K L HD HQC G F K A F R R DL I L D L G K E V - K D NHWF WD T E V 239 Aper 93 I L DA DI P V R P I F I NQA V V L A MNL G I DL I I A NR V Y ------R T HS L L R R V L S V A Y NS L V NL L F K - - T G L R D HQA G L K I L S R R A A K I I L MK R T R T DG L A Y D T E I 186 Hmar 95 Y F DT DL A T DMS HL E E L V NA V R V DG Y DV A T G S R W- - L P E NR A DR P A K R G I P S F G Y NT L V R T V L R - - S DL K D HQC G F K A F D R G A L E T L L P L V - QDE HWF WDT E L 191 HNRC 97 Y F DT DL A T DMR HL E T L V E R V R T G S A DV A T G S R W- - MP G E T A DR P A K R G I P S R V F NG A V R T L L G - - S S V R DHQC G F K A L S R S A F E A L V D D V - A DE HWF WD T E L 193 Hwa1 133 Y F DT DL A T DI R HL E E L I T R I QT G E A DI A T G S R W- - L P E NI A DR P A K R G I P S R V Y NT L V R L F L R - - S D L R DHQC G F K A F S R E A F E S L QP I V - E DS HWF WD T E M 229 Mbar 87 Y I DV DL A T DMK Y L E K L I R A V S T DG Y DF A T G S R M- - MP DS DA K R P F K R E F A S R G Y NF L V R L F L H- - S K L Y D HQC G F K A F R R E A L F E L S E D V - E NE HWF WDT E V 182 Mhun 88 F MDA DNS T K V S E L V R L S R R I G - - DHDG V I G S R HL P G QV L QR K QP L F R R I QS R I F NG L I R L L F G - - L P F Y DT QC G A K I F K K QA L DA V L P HL - R S T G F E F DV E L 184 Mkan 91 C MDA DG QHP P E C L P NI V NP V L DG E C DF G L G S R Y - V E G S V V E NF P WY R K L NS WG A R V V A R L F L K - - L P Y R DP T S G F R A I S R K I L T E S R P - F - V S E G F E I QV E T 187 Msed 80 F L DA DL P V G K E D L MR V I QE A R - - DHDL V I T T R I - - F R N- - - - MP T NR S F L HR A F V S V A K V F F P S L S F V R DF QS G L K V A R R E K L L QV K DE L - V MS D WL F D V NL 172 Npha 95 Y F DT DL A T DMR HL E A L V E S V R T E G Y DI A T G S R R - - MP G K R QR R E P E R G I A S T G Y NA L V R L F L R - - S P L Y D HQC G F K A F D R DA L L A L A D D I - E DNHWF WDT E L 191 Paby 90 MMDP DG S Y DP K E I P K L L E I L R K E A A DF V I G S R L K G K I E P G A MP WL HR Y I G NP L L T K I L NF L F K - - I K V S D A HS G F R A I K R DA L QK L T - - L - K C R G ME F A S E M 186 Ptor 82 Y MDA DL S HR P E D I K G MI E K A I K T NA DL V I G S R Y - - I DNG E T HDE F I R QI I S K T A NR L F R L S F N- - L NV HDC T S G F R I Y S R R A C DF L A R QV D I E NG Y V G QI DI 179 Ssol 102 L L DA DF P I T E E E L NK I L S T - - - - DA DL V I P R R K - - I I G - - - - MP L K R R F L HK A F I V L T K I L F P S L L K F S DF QA G V K L V NR E K V V S V L DE L - I I ND F L F D V NL 192 Tkod 88 F MDA DG QHL P E E I G K L V R P I V E G R A DL V I G A R - - - K V E V QG K R P L HR R L S NI I T T R L I R L K L G - - T Y V Y D T QS G F R A Y R R G F L P E I E - - - - - S D R Y E V E T E M 179 Tvol 103 L MDA DG S V P L K E I V K A L DL T N- - Y Y DL I I F D R Y - - - S NR G NR I P F I R R F P S R G F NK L V R I F L G - - L K I ND T QC G Y K I I K R E Y A QR A F NK I - T I S NA F F D V A L 196

Figure 2: Conserved residues in archaeal AglD proteins. The sequences of Hfx. Volcanii AglD (accession number AM698042) and AglD homologues in Archaeoglobus fulgidus (NP 069415.1; Aful), Aeropyrum pernix (NP 147774.1; Aper), Haloarcula marismortui (YP 136461.1; Hmar), Halobacterium sp. NRC-1 (NP 279416.1; HNRC), Haloquadratum walsbyi (YP 657261.1; Hwal), Methanosarcina acetivorans (NP 618739.1; Mace), Methanosarcina barkeri (YP 304067.1; Mbar), Methanospirillum hungatei (YP 503949.1; Mhun), Methanopyrus kandleri (NP 614163.1; Mkan), Metallosphaera sedula (YP 001191894; Msed), Natronomonas pharaonis (YP 326773.1; Npha), Pyrococcus abyssi (NP 127133.1; Paby), Picrophilus torridus (YP 024256.1; Ptor), Sulfolobus solfataricus (NP 342803.1; Ssol), Thermococcus kodakarensis (YP 182777.1; Tkod), and Thermoplasma volcanium (NP 111403.1; Tvol) were aligned by ClustalW2 (www.ebi.ac.uk/Tools/clustalw2/index.html), using the default settings. The region of the highest similarity is shown. Completely conserved residues are shown against a black background, while largely conserved residues (i.e., similar residues conserved in at least 11 sequences) are shown against a grey background. Amino acid numbers are shown at the start and end of each sequence. Asterisks are placed under Hfx. volcanii AglD D110, D112, R139, D173, D195, and D201 (see text for details).

(Asp110) (Asp112) mutant mutant aglD aglD aglD aglD / / / / Mutation Mutation aglD aglD aglD aglD aglD aglD Δ Δ bkgnd Δ Δ bkgnd bkgnd Δ Δ bkgnd

CBB CBB

D110A D112N PAS PAS

CBB CBB

D110E D112E PAS PAS

Figure 3: Hfx. volcanii AglD Asp110 and Asp112 residues likely participate in the GT2 DXD motif involved in the catalytic activity of the enzyme. Site-directed mutagenesis was performed to generate CBD-AglD containing mutations of Asp110 (left column) or Asp112 (right column), as listed on the left of each panel. For each mutant, the upper and lower panels, respectively, show the Coomassie- and PAS- stained S-layer glycoprotein from the background strain (lanes 1 and 5), from the aglD deletion strain (lane 2), from the aglD deletion strain complemented with a plasmid encoding CBD-AglD (lane 3), or from the aglD deletion strain complemented with a plasmid encoding CBD fusedtomutatedAglD(lane4).

Asp201 for AglD activity was considered by site-directed were fully restored in cells expressing AglD D173N, it would mutagenesis. appear that the conserved Asp173 does not serve as the Transformation of Hfx. volcanii ΔaglD cells with a catalytic base of the enzyme but is likely important for AglD plasmid encoding AglD D173E (Figure 4) did not restore structure. This idea may be supported by the observation the SDS-PAGE migration or glycostaining of the S-layer that the D173A mutant could not be expressed (not shown). glycoprotein. However, as these S-layer glycoprotein traits The importance of Asp195, another somewhat conserved Archaea 5

(Asp201) mutant aglD aglD / / aglD Mutation aglD aglD Δ (Asp173) bkgnd Δ Δ bkgnd CBB

mutant D201A aglD / PAS Mutation aglD aglD/aglD aglD bkgnd Δ Δ Δ bkgnd (Asp195)

CBB CBB

D173E D201E mutant aglD aglD / PAS PAS / aglD Mutation aglD aglD Δ bkgnd Δ Δ bkgnd

CBB CBB CBB

D173N D201N D195A PAS PAS PAS

Figure 4: Hfx. volcanii AglD Asp201 is apparently the catalytic base of the enzyme. Site-directed mutagenesis was performed to generate CBD-AglD containing mutations of Asp173 (upper left column), Asp195 (lower left column), and Asp201 (right column), as listed on the left of each panel. For each mutant, the upper and lower panels, respectively, show the Coomassie- and PAS-stained S-layer glycoprotein from the background strain (lanes 1 and 5), from the aglD deletion strain (lane 2), from the aglD deletion strain complemented with a plasmid encoding CBD-AglD (lane 3), or from the aglD deletion strain complemented with a plasmid encoding CBD fused to mutated AglD (lane 4).

Asp residue in this region, was also considered. Hfx. volcanii 3.5. The Conserved Arg139 Residue Is Needed for AglD Activity. ΔaglD cells transformed to express AglD D195A (Figure 4) Hfx. volcanii AglD and its archaeal homologues also contain or D195E (not shown) readily replaced the actions of the several other fully conserved residues in that part of the missing enzyme, showing that the Asp at this position is soluble N-terminal region under consideration in this study. not necessary for AglD activity. By contrast, if the aglD- The contribution of these residues, as well as that of their deleted strain was transformed to express AglD D201A or neighbors, was next considered. Complementation of ΔaglD D201N, SDS-PAGE migration and glycostaining of the S- Hfx. volcanii cells with plasmid-encoded AglD R139A failed layer glycoprotein were as observed in cells lacking native to restore either S-layer glycoprotein migration in SDS- AglD (Figure 4). When, however, Hfx. volcanii ΔaglD cells PAGE or the ability of PAS glycostain to label this reporter were transformed to express the D201E mutant, the S- (Figure 6). The same was true in cells expressing AglD layer glycoprotein behaved as in the background strain R139E, R139K or R139M (not shown). Thus, AglD Arg139 (Figure 4). These results thus point to Asp201 as being is apparently essential for enzyme activity. By contrast, the catalytic base of AglD and the functional equivalent complementation of AglD-lacking Hfx. volcanii cells with of SpsA Asp191. Furthermore, as is the case with other plasmid-encoded AglD D133A or G177A restored S-layer GT2 family members [22], Asp201 can be replaced by glycoprotein SDS-PAGE migration to that observed for the Glu. native protein, although less S-layer glycoprotein is detected Homology modeling of Hfx. volcanii AglD residues in the cells expressing AglD D133A. The signiﬁcance of this Asp110-Asp112, Asp173, Asp195, and Asp201, based on the observation is not clear. In addition, the S-layer glycoprotein available three-dimensional structural of B. subtilis SpsA in both AglD D133A- and G177A-expressing cells could [20], further supports the assignment of AglD D201 as be glycostained (Figure 6). As such, although conserved in being equivalent to subtilis SpsA Asp191. As reﬂected in Hfx. volcanii AglD and its archaeal homologues, Asp133 Figure 5, considerable overlap in term of both position and and Gly177 do not appear to be essential for the catalytic orientation exists between AglD D201 and B. subtilis SpsA workings of the enzyme. Similarly, the introduction of Asp191. The same cannot be said for either AglD Asp173 or CBD-tagged AglD G137A, S138A, Q175A, C176A, F178A Asp195. or K179A mutants into Hfx. volcanii ΔaglD cells led to a 6 Archaea

Asp195

Asp110 Asp191 Thr111 Asp201 Asp112 Thr97 Asp98 Asp99

Asp173

Figure 5: Homology modeling of Hfx. volcanii AglD residues based on the available three-dimensional structure of B. subtilis SpsA. Structural modeling was performed by using the SWISS-MODEL program (http://swissmodel.expasy.org/) and visualized using PyMol (http://www.pymol.org/). B. subtilis SpsA Thr97, Asp98, Asp99, and Asp191 are shown in blue, while Hfx. volcanii AglD Asp110, Thr111, Asp112, Asp173, Asp195, and Asp 201 are shown in brown. The ribbon diagram in the background corresponds to the three-dimensional structure of SpsA [20]. The RMS value, reflecting the quality of the homology modeling, is 0.61 angstroms. restoration of AglD activity, indicating that none of these only demonstrated to modify an artificial substrate [7, 35]. residues contribute to the reaction catalyzed by the enzyme Similarly, while both Thermoplasma acidophilum [36]and (not shown). Pyrococcus horikoshii [37]havebeenreportedtocontain Finally, to eliminate the possibility that the inability of glycoproteins, the participation of biochemically character- certain plasmid-encoded versions of the protein to restore ized dolichyl phosphomannose synthases from these species absent AglD activity was due to poor or no expression, the [37, 38] in protein glycosylation has yet to be demonstrated. level of each CBD-AglD considered in this study was assessed As such, the present analysis of Hfx. volcanii AglD represents by immunoblot using antiCBD antibodies (Figure 7). the first examination of a glycosyltransferase experimentally verified as participating in the modification of an identified archaeal glycoprotein, namely, the S-layer glycoprotein. 4. Discussion In the present study, sequence alignment was combined When one considers that Nanoarchaeum equitans, the with site-directed mutagenesis to identify AglD residues archaeon containing the smallest genome identified to date important for the function of the enzyme, as reflected in [31, 32], encodes only 3 GTs, namely, one member of the AglD-mediated modulation of the SDS-PAGE migration GT2familyandtwomembersoftheGT4family[33], it and glycostaining of a reporter glycoprotein, the S-layer is fair to say that analysis of archaeal GTs can provide glycoprotein. This approach assigned AglD Asp110, Thr111, unique insight into the evolution of such enzymes, as well and Asp112 as the DXD motif typical of inverting GT-A fold- as adding to our comprehension of protein processing in bearing GTs. AglD Asp110 was shown to be essential for extreme conditions. Despite such promise, only limited catalytic activity, while a negative charge at position 112 was experimental data on archaeal GTs involved in protein deemed necessary. In the case of the AglD D112E mutant, glycosylation is presently available. The crystal structure of recovery of S-layer glycoprotein SDS-PAGE migration was Stt3/AglB from Pyrococcus furiosus, the sole component of noted, yet the loss of PAS glycostaining associated with the the archaeal oligosaccharyltransferase [5, 6], has been solved deletion strain was largely not restored. This could reflect [7], shedding new light on the workings of this central the generation of an enzyme possessing different activity component of the N-glycosylation machinery. Still, although than that of the native protein, one that adds a different P. furiosus has been reported to contain glycoproteins [34], sugar to the final position of S-layer glycoprotein-bound the oligosaccharyltransferase in this species has been thus far pentasaccharide. Indeed, the failure of PAS glycostain to label Archaea 7

mutant Mutation: Wild type D110A D110E D110N D112E D112N aglD aglD / / aglD Mutation aglD aglD Δ bkgnd Δ Δ bkgnd

CBB

R139A D133A G137A S138A R139A R139E R139K R139M

PAS

CBB R152K D173E D173N Q175A C176A D133A

PAS F178A K179A D195A D195E D201A D201E D201N CBB

G177A Figure 7: Expression levels of the various versions of CBD-AglD. PAS Hfx. volcanii cells expressing the various AglD mutants considered in this study were grown to OD550 1.0 and their protein contents were separated on 10% SDS-PAGE. The CBD-AglD content of each Figure 6: The conserved Arg139 residue is needed for AglD activity, strain was subsequently assessed by immunoblot using polyclonal unlike the conserved Asp133 or Gly177 residues. Site-directed antiCBD antibodies. Antibody binding was detected using HRP- mutagenesis was performed to generate CBD-AglD containing conjugated secondary antibodies and an enhanced chemilumines- mutations of Asp133, Arg139, and Gly177, as listed on the left of cence kit. each panel. For each mutant, the upper and lower panels respectively show the Coomassie- and PAS-stained S-layer glycoprotein from the background strain (lanes 1 and 5), from the aglD deletion strain (lane 2), from the aglD deletion strain complemented with a plasmid encoding CBD-AglD (lane 3), or from the aglD deletion Asp187 catalytic base in S. mililoti ExoM led to a complete strain complemented with a plasmid encoding CBD fused to loss of function [21]. Such nuances may be indicative of mutated AglD (lane 4). differences in the donors and/or acceptors employed by each enzyme or point to unique mechanistic traits. In addition, although conserved in the archaeal AglD homologues the tetrasaccharide N-linked to the S-layer glycoprotein in examined in this study, Hfx. volcanii AglD Asp173 was cells lacking AglD points to inability of this labeling reagent not assigned as the catalytic base of the enzyme, given its to interact with certain sugar subunits. Moreover, within the functional replacement by a similarly sized Asn but not a GT2 family (whose members include Hfx. volcanii AglD), similarly charged Glu. Hence, it would appear that Asp173 differences in the organization and importance of DXD motif is of structural, rather than catalytic, importance to AglD constituents exist. In the case of S. meliloti ExoM, where activity. In addition to these residues, at least another Hfx. the DXD motif includes Asp96 and Asp98, it was shown volcanii AglD amino acid seems to be needed for enzyme that replacing the former with Ala completely eliminated function, that is, Arg139. The AglD counterparts of other enzymatic activity, whereas the same replacement at position residues shown to be important for the catalytic activity of 98 only led to a 70% loss of activity [21]. In S. enterica GT2 family members, such as S. meliloti ExoM Asp44 and WbbE, where the DXD motif is expanded to include Asp93, Asp96 [21], may also play a role in the activity of the archaeal Asp95, and Asp96, it was shown that exchanging either enzyme. Asp93 or Asp96 with Ala abolished enzyme activity, while In conclusion, this paper describes an in vivo assay the same replacement at Asp95 only reduced that activity designed to consider the contribution of various AglD [22]. residues to the activity of the enzyme. In the assay, the ability The site-directed mutagenesis approach developed here, of plasmid-encoded versions of AglD, selectively mutated at along with sequence alignment and homology modeling, positions suspected of being important for enzyme function, also indicate Asp201 as likely serving as the AglD catalytic to restore both S-layer glycoprotein SDS-PAGE migration base. Just as the corresponding residue in S. enterica WbbE, and glycostaining affected in ΔaglD cells is assessed. In this that is, Glu180, could be functionally replaced by Asp [22], manner, Asp110, Asp112, and Asp201 were all determined as AglD D201E was also active. By contrast, replacing the being important for AglD activity, as was Asp139. 8 Archaea

Acknowledgment [14] C. A. Wiggins and S. Munro, “Activity of the yeast MNN1 alpha-1,3-mannosyltransferase requires a motif conserved in The authors thank Dr. Raz Zarivach for assistance with the many other families of glycosyltransferases,” Proceedings of the homology modeling. Support came from the Israel Science National Academy of Sciences of the United States of America, Foundation (Grant 30/07) and the US Army Research Office vol. 95, pp. 7945–7950, 1998. (Grant W911NF-07-1-0260). [15] C. Breton and A. Imberty, “Structure/function studies of glycosyltransferases,” Current Opinion in Structural Biology, vol. 9, no. 5, pp. 563–571, 1999. [16] U. M. Unligil and J. M. Rini, “Glycosyltransferase structure References and mechanism,” Current Opinion in Structural Biology, vol. [1] M. F. Mescher and J. L. Strominger, “Purification and 10, no. 5, pp. 510–517, 2000. characterization of a prokaryotic glycoprotein from the cell [17] C. Breton, L. Snajdrovˇ a,´ C. Jeanneau, J. Koca,ˇ and A. envelope of Halobacterium salinarium,” The Journal of Biolog- Imberty, “Structures and mechanisms of glycosyltransferases,” ical Chemistry, vol. 251, no. 7, pp. 2005–2014, 1976. Glycobiology, vol. 16, no. 2, pp. 29R–37R, 2006. [2] M. Abu-Qarn, J. Eichler, and N. Sharon, “Not just for Eukarya [18] J. A. Campbell, G. J. Davies, V. Bulone, and B. Henrissat, anymore: N-glycosylation in Bacteria and Archaea,” Current “A classification of nucleotide-diphospho-sugar glycosyltrans- Opinion in Structural Biology, vol. 18, no. 5, pp. 544–550, 2008. ferases based on amino acid sequence similarities,” Biochemi- [3]S.Yurist-Doutsch,B.Chaban,D.J.VanDyke,K.F.Jarrell, cal Journal, vol. 326, no. 3, pp. 929–939, 1997. and J. Eichler, “Sweet to the extreme: protein glycosylation [19] I. M. Saxena, R. M. Brown Jr., M. Fevre, R. A. Geremia, in Archaea,” Molecular Microbiology, vol. 68, no. 5, pp. 1079– and B. Henrissat, “Multidomain architecture of β-glycosyl 1084, 2008. transferases: implications for mechanism of action,” Journal of [4]D.J.VanDyke,J.Wu,andS.M.Logan,“Identificationofgenes Bacteriology, vol. 177, no. 6, pp. 1419–1424, 1995. involved in the assembly and attachment of a novel flagellin N- [20] S. J. Charnock and G. J. Davies, “Structure of the nucleotide- linked tetrasaccharide important for motility in the archaeon diphospho-sugar transferase, SpsA from Bacillus subtilis, in Methanococcus maripaludis,” Molecular Microbiology, vol. 72, native and nucleotide-complexed forms,” Biochemistry, vol. no. 3, pp. 633–644, 2009. 38, no. 20, pp. 6380–6385, 1999. [5] M. Abu-Qarn and J. Eichler, “Protein N-glycosylation in [21] C. Garinot-Schneider, A. C. Lellouch, and R. A. Geremia, Archaea: defining Haloferax volcanii genes involved in S-layer “Identification of essential amino acid residues in the Sinorhi- glycoprotein glycosylation,” Molecular Microbiology, vol. 61, zobium meliloti glucosyltransferase ExoM,” The Journal of pp. 511–525, 2006. Biological Chemistry, vol. 275, no. 40, pp. 31407–31413, 2000. [6] B. Chaban, S. Voisin, J. Kelly, S. M. Logan, and K. F. Jarrell, [22] W. J. Keenleyside, A. J. Clarke, and C. Whitfield, “Identifica- “Identification of genes involved in the biosynthesis and tion of residues involved in catalytic activity of the inverting attachment of Methanococcus voltae N-linked glycans: insight glycosyl transferase WbbE from Salmonella enterica serovar into N-linked glycosylation pathways in Archaea,” Molecular borreze,” Journal of Bacteriology, vol. 183, no. 1, pp. 77–85, Microbiology, vol. 61, no. 1, pp. 259–268, 2006. 2001. [7] M. Igura, N. Maita, J. Kamishikiryo, et al., “Structure-guided [23] B. W. Murray, S. Takayama, J. Schultz, and C.-H. Wong, identification of a new catalytic motif of oligosaccharyltrans- “Mechanism and specificity of human α-1,3-fucosyl- ferase,” EMBO Journal, vol. 27, no. 1, pp. 234–243, 2008. transferase V,” Biochemistry, vol. 35, no. 34, pp. 11183–11195, [8] N. Taniguchi, A. Ekuni, J. H. Ko, et al., “A glycomic approach 1996. to the identification and characterization of glycoprotein [24] L. C. Pedersen, T. A. Darden, and M. Negishi, “Crystal function in cells transfected with glycosyltransferase genes,” structure of β1,3-glucuronyltransferase I in complex with Proteomics, vol. 1, no. 2, pp. 239–247, 2001. active donor substrate UDP-GlcUA,” The Journal of Biological [9] P. M. Coutinho, E. Deleury, G. J. Davies, and B. Henrissat, “An Chemistry, vol. 277, no. 24, pp. 21869–21873, 2002. evolving hierarchical family classification for glycosyltrans- [25] S. Kakuda, T. Shiba, M. Ishiguro, et al., “Structural basis ferases,” Journal of Molecular Biology, vol. 328, no. 2, pp. 307– foracceptorsubstraterecognitionofahumanglucuronyl- 317, 2003. transferase, GlcAT-P, an enzyme critical in the biosynthesis [10] B. I. Cantarel, P. M. Coutinho, C. Rancurel, T. Bernard, ofthecarbohydrateepitopeHNK-1,”The Journal of Biological V. Lombard, and B. Henrissat, “The Carbohydrate-Active Chemistry, vol. 279, no. 21, pp. 22693–22703, 2004. EnZymes database (CAZy): an expert resource for glycoge- [26] M. Abu-Qarn, S. Yurist-Doutsch, A. Giordano, et al., nomics,” Nucleic Acids Research, vol. 37, supplement 1, pp. “Haloferax volcanii AglB and AglD are involved in N- D233–D238, 2009. glycosylation of the S-layer glycoprotein and proper assembly [11] L. L. Lairson, B. Henrissat, G. J. Davies, and S. G. Withers, of the surface layer,” Journal of Molecular Biology, vol. 374, no. “Glycosyl transferases: structures, functions, and mecha- 5, pp. 1224–1236, 2007. nisms,” Annual Review of Biochemistry, vol. 77, pp. 521–555, [27] H. Magidovich, S. Yurist-Doutsch, Z. Konrad, et al., “AglP 2008. is a S-adenosyl-L-methionine-dependent methyltransferase [12] B. Henrissat, G. Sulzenbacher, and Y. Bourne, “Glycosyl- that participates in the N-glycosylation pathway of Haloferax transferases, glycoside hydrolases: surprise, surprise!,” Current volcanii,” Molecular Microbiology, vol. 76, pp. 190–199, 2010. Opinion in Structural Biology, vol. 18, no. 5, pp. 527–533, 2008. [28] M. Mevarech and R. Werczberger, “Genetic transfer in [13]C.Breton,E.Bettler,D.H.Joziasse,R.A.Geremia,andA. Halobacterium volcanii,” Journal of Bacteriology, vol. 162, pp. Imberty, “Sequence-function relationships of prokaryotic and 461–462, 1985. eukaryotic galactosyltransferases,” Journal of Biochemistry, vol. [29] N. Plavner and J. Eichler, “Defining the topology of the N- 123, no. 6, pp. 1000–1009, 1998. glycosylation pathway in the halophilic archaeon Haloferax Archaea 9

volcanii,” Journal of Bacteriology, vol. 190, no. 24, pp. 8045– 8052, 2008. [30] G. Dubray and G. Bezard, “A highly sensitive periodic acid- silver stain for 1,2-diol groups of glycoproteins and polysaccharides in polyacrylamide gels,” Analytical Biochemistry, vol. 119, no. 2, pp. 325–329, 1982. [31]H.Huber,M.J.Hohn,R.Rachel,T.Fuchs,V.C.Wimmer, and K. O. Stetter, “A new phylum of Archaea represented by a nanosized hyperthermophilic symbiont,” Nature, vol. 417, no. 6884, pp. 63–67, 2002. [32] E. Waters, M. J. Hohn, and I. Ahel, “The genome of Nanoarchaeum equitans: insights into early archaeal evolution and derived parasitism,” Proceedings of the National Academy of Sciences of the United States of America, vol. 100, pp. 12984– 12988, 2003. [33] H. Magidovich and J. Eichler, “Glycosyltransferases and oligosaccharyltransferases in Archaea: putative components of the N-glycosylation pathway in the third domain of life,” FEMS Microbiology Letters, vol. 300, pp. 120–130, 2009. [34] M. V. Weinberg, G. J. Schut, S. Brehm, S. Datta, and M. W. W. Adams, “Cold shock of a hyperthermophilic archaeon: Pyrococcus furiosus exhibits multiple responses to a suboptimal growth temperature with a key role for membrane-bound glycoproteins,” JournalofBacteriology, vol. 187, no. 1, pp. 336– 348, 2005. [35] D. Kohda, M. Yamada, M. Igura, J Kamishikiryo, and K. Maenaka, “New oligosaccharyltransferase assay method,” Gly- cobiology, vol. 17, pp. 1175–1182, 2007. [36] L. L. Yang and A. Haug, “Puriﬁcation and partial characterization of a procaryotic glycoprotein from the plasma membrane of Thermoplasma acidophilum,” Biochimica et Biophysica Acta, vol. 556, no. 2, pp. 265–277, 1979. [37] Y. Urushibata, S. Ebisu, and I. Matsui, “A thermostable dolichol phosphoryl mannose synthase responsible for gly- coconjugate synthesis of the hyperthermophilic archaeon Pyrococcus horikoshii,” Extremophiles, vol. 12, no. 5, pp. 665– 676, 2008. [38] B. C. R. Zhu and R. A. Laine, “Dolichyl-phosphomannose synthase from the Archae Thermoplasma acidophilum,” Gly- cobiology, vol. 6, no. 8, pp. 811–816, 1996.

Chapter 3.4

"AglR is required for addition of the final mannose residue of the N-

linked glycan decorating the Haloferax volcanii

S-layer glycoprotein"

Lina Kaminski, Ziqiang Guan, Mehtap Abu-Qarn, Zvia Konrad and Jerry Eichler Biochim. Biophys. Acta. 2012. 1820:1664-1670

Abstract:

Background: Recent studies of Haloferax volcanii have begun to elucidate the steps of N-glycosylation in Archaea, where this universal post-translational modification remains poorly described. In Hfx. volcanii, a series of Agl proteins catalyzes the assembly and attachment of a N-linked pentasaccharide to the S-layer glycoprotein. Although roles have been assigned to the majority of Agl proteins, others await description. In the following, the contribution of AglR to N-glycosylation was addressed.

Methods: A combination of bioinformatics, gene deletion, mass spectrometry and metabolic radiolabeling served to show a role for AglR in archaeal N- glycosylation at both the dolichol phosphate and reporter glycoprotein levels.

Results: The modified behavior of the S-layer glycoprotein isolated from cells lacking AglR points to an involvement of this protein in N-glycosylation. In cells lacking AglR, glycan-charged dolichol phosphate, including mannose-charged dolichol phosphate, accumulates. At the same time, the S-layer glycoprotein does not incorporate mannose, the final subunit of the N-linked pentasaccharide decorating this protein. AglR is a homologue of Wzx proteins, annotated as flippases responsible for delivering lipid-linked O-antigen precursor oligosaccharides across the bacterial plasma membrane during lipopolysaccharide biogenesis.

Conclusions: The effects resulting from aglR deletion are consistent with AglR interacting with dolichol phosphate-mannose, possibly acting as a dolichol phosphate- mannose flippase or contributing to such activity.

General significance: Little is known of how lipid-linked oligosaccharides are translocated across membrane during N-glycosylation. The possibility of Hfx. volcanii

AglR mediating or contributing to flippase activity could help address this situation.

Biochimica et Biophysica Acta 1820 (2012) 1664–1670

Contents lists available at SciVerse ScienceDirect

Biochimica et Biophysica Acta

journal homepage: www.elsevier.com/locate/bbagen

AglR is required for addition of the ﬁnal mannose residue of the N-linked glycan decorating the Haloferax volcanii S-layer glycoprotein

Lina Kaminski a,1, Ziqiang Guan b,1, Mehtap Abu-Qarn a, Zvia Konrad a, Jerry Eichler a,⁎ a Department of Life Sciences, Ben Gurion University of the Negev, Beersheva 84105, Israel b Department of Biochemistry, Duke University Medical Center, Durham, NC 27710, USA article info abstract

Article history: Background: Recent studies of Haloferax volcanii have begun to elucidate the steps of N-glycosylation in Received 17 April 2012 Archaea, where this universal post-translational modification remains poorly described. In Hfx. volcanii,a Received in revised form 15 June 2012 series of Agl proteins catalyzes the assembly and attachment of a N-linked pentasaccharide to the S-layer gly- Accepted 18 June 2012 coprotein. Although roles have been assigned to the majority of Agl proteins, others await description. In the Available online 27 June 2012 following, the contribution of AglR to N-glycosylation was addressed. Methods: A combination of bioinformatics, gene deletion, mass spectrometry and metabolic radiolabeling Keywords: served to show a role for AglR in archaeal N-glycosylation at both the dolichol phosphate and reporter glyco- Archaea Dolichylphosphate-mannose protein levels. Haloferax volcanii Results: The modified behavior of the S-layer glycoprotein isolated from cells lacking AglR points to an in- N-glycosylation volvement of this protein in N-glycosylation. In cells lacking AglR, glycan-charged dolichol phosphate, includ- S-layer glycoprotein ing mannose-charged dolichol phosphate, accumulates. At the same time, the S-layer glycoprotein does not incorporate mannose, the final subunit of the N-linked pentasaccharide decorating this protein. AglR is a homologue of Wzx proteins, annotated as flippases responsible for delivering lipid-linked O-antigen precursor oligosaccharides across the bacterial plasma membrane during lipopolysaccharide biogenesis. Conclusions: The effects resulting from aglR deletion are consistent with AglR interacting with dolichol phosphate-mannose, possibly acting as a dolichol phosphate-mannose flippase or contributing to such activity. General significance: Little is known of how lipid-linked oligosaccharides are translocated across membrane during N-glycosylation. The possibility of Hfx. volcanii AglR mediating or contributing to flippase activity could help address this situation. © 2012 Elsevier B.V. All rights reserved.

1. Introduction years, however, studies on the halophilic archaeon, Haloferax volcanii, have offered significant insight into N-glycosylation in Archaea. Members of all three domains of life, i.e. Eukarya, Bacteria and In Hfx. volcanii, select Asn residues of the S-layer glycoprotein, a Archaea, perform N-glycosylation, namely the covalent attachment reporter of N-glycosylation in this species, are decorated with a of glycans to select Asn residues of target proteins. Whereas the path- pentasaccharide comprising a hexose, two hexuronic acids, a methyl ways of this post-translational modification in Eukarya and Bacteria are ester of hexuronic acid and a mannose [5–7]. A series of Agl (archaeal relatively well defined, much regarding the archaeal version of this uni- glycosylation) proteins assembles these sugar residues eventually versal protein-processing event remains unknown [1–4].Inrecent comprising the N-linked pentasaccharide onto dolichol phosphate (DolP) carriers [7].Specifically, AglJ, AglG, AglI, and AglE sequentially add the first four pentasaccharide residues onto a common DolP, while AglD adds the final pentasaccharide residue, mannose, to a distinct DolP [5,8–10]. The DolP-bound tetrasaccharide (and its precursors) is trans- Abbreviations: ABC, ATP-binding cassette; CBB, Coomassie brilliant blue; ferred to select S-layer glycoprotein Asn residues by the archaeal oli- DDW, double-distilled water; DolP, dolichol phosphate; DolPP-Man5, mannose - 5 gosaccharyltransferase, AglB [5].Mannose,thefinal pentasaccharide N-acetylglucosamine2-charged dolichol pyrophosphate; EIC, extracted ion chromatograms; ER, endoplasmic reticulum; LC-ESI/MS, liquid chromatography-electrospray ionization residue, is subsequently transferred from it DolP carrier to the N-linked mass spectrometry; LLO, lipid-linked oligosaccharide; Man, mannose; MS/MS, tandem tetrasaccharide [7,11]. Other Agl proteins, such as AglF, AglM and AglP, mass spectrometry; RT-PCR, reverse transcriptase-PCR serve sugar-processing roles important for N-glycosylation. ⁎ Corresponding author at: Dept. of Life Sciences, Ben Gurion University of the Despite these advances in delineating the Hfx. volcanii N-glycosylation Negev, PO Box 653, Beersheva 84105, Israel. Tel.: +972 8646 1343; fax: +972 8647 9175. E-mail address: [email protected] (J. Eichler). pathway, enzymes catalyzing several steps of the process remain to be 1 These authors contributed equally to this work. identified. For instance, it is not clear how mannose finds its way

0304-4165/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.bbagen.2012.06.014 L. Kaminski et al. / Biochimica et Biophysica Acta 1820 (2012) 1664–1670 1665 from DolP to the N-linked tetrasaccharide decorating the S-layer analysis of the dolichol phosphate pool, the total lipid extracts were sub- glycoprotein. In Hfx. volcanii, genes encoding the Agl proteins involved jected to normal phase LC/MS analysis without pre-fractionation. in N-glycosylation (apart from aglD) are found within a single cluster Normal phase LC-ESI/MS of lipids was performed using an Agilent [12]. While many of the products of the agl gene clusters have been 1200 Quaternary LC system coupled to a QSTAR XL quadrupole demonstrated as participating in N-glycosylation, in other cases, evi- time-of-flight tandem mass spectrometer (Applied Biosystems, Foster dence for such involvement has yet to be provided. For instance, while City, CA). An Ascentis Si HPLC column (5 μm, 25 cm×2.1 mm) was aglR is co-transcribed with aglE [12], the requirement for AglR in used. Mobile phase A consisted of chloroform/methanol/aqueous ammo- N-glycosylation has yet to be addressed. In the following, an analysis nium hydroxide (800:195:5, v/v/v). Mobile phase B consisted of chloro- of the effects of aglR deletion points to AglR serving a role in form/methanol/water/aqueous ammonium hydroxide (600:340:50:5, DolP-mannose processing, possibly mediating or contributing to Hfx. v/v/v/v). Mobile phase C consisted of chloroform/methanol/water/ volcanii DolP-mannose flippase activity. aqueous ammonium hydroxide (450:450:95:5, v/v/v/v). The elution program consisted of the following: 100% mobile phase A was held 2. Materials and methods isocratically for 2 min and then linearly increased to 100% mobile phase B over 14 min and held at 100% B for 11 min. The LC gradient 2.1. Strains and growth conditions was then changed to 100% mobile phase C over 3 min and held at 100% C for 3 min, and finally returned to 100% A over 0.5 min and fl μ The Hfx. volcanii parent H53 strain and the isogenic strains deleted of held at 100% A for 5 min. The total LC ow rate was 300 l/min. The post-column splitter diverted ~10% of the LC flow to the ESI aglR were grown in medium containing 3.4 M NaCl, 0.15 M MgSO4, 1 mM MnCl , 4 mM KCl, 3 mM CaCl , 0.3% (w/v) yeast extract, 0.5% source of the Q-Star XL mass spectrometer, with MS settings as 2 2 − (w/v) tryptone, 50 mM Tris–HCl, pH 7.2, at 40 °C [13].TheHfx. volcanii follows: ion spray voltage= 4500 V, curtain gas=20 psi, ion − strains deleted of aglB and aglD were previously described [14]. source gas 1=20 psi, de-clustering potential= 55 V and focusing potential=−150 V. Nitrogen was used as the collision gas for MS/ MS experiments. Data acquisition and analysis were performed 2.2. Deletion of aglR using the instrument's Analyst QS software. LC-ESI/MS/MS analysis of S-layer glycoprotein tryptic fragments Deletion of aglR was performed as previously described [14,15].To was performed as previously described [16]. The protein contents of fl amplify approximately 500 bp-long regions anking the coding sequence Hfx. volcanii cells were separated on 7.5% polyacrylamide gels and ′ of aglR,theaglR-5upfor (gggctcgagGGAGTTCATCAATATGGTCCG; stained with Coomassie R-250 (Fluka, St. Louis MO). For in-gel digestion ′ genomic sequence in capital letters) and aglR-5 uprev (cccaagcttTC of the S-layer glycoprotein, the relevant bands (identified via the unique GTGTTATTACCGCACCACGGA) primers, directed against the upstream SDS-PAGE migration and staining pattern of the protein) were excised, fl ′ anking region, and the aglR-3 downfor (ggggatccGGTATGACTAGG destained in 400 μl of 50% (vol/vol) acetonitrile (Sigma, St Louis, MO) in ′ TCGCAAGTA) and aglR-3 downrev (ccctctagaATGCTCTCTTTCATTTGCA 40 mM NH HCO , pH 8.4, dehydrated with 100% acetonitrile, and dried fl 4 3 TATT) primers, directed against the downstream anking region, were using a SpeedVac drying apparatus. The S-layer glycoprotein was ′ employed. XhoIandHindIII sites were introduced using the aglR-5 reduced with 10 mM dithiothreitol (Sigma) in 40 mM NH HCO at ′ 4 3 upfor and aglR-5 uprev primers, respectively, while BamHI and XbaI 56 °C for 60 min and then alkylated for 45 min at room temperature sites were introduced using the aglR-3′downfor and aglR-3′downrev with 55 mM iodoacetamide in 40 mM NH4HCO3. The gel pieces were primers, respectively. washed with 40 mM NH HCO for 15 min, dehydrated with 100% fi fi 4 3 To con rm deletion of aglR at the DNA level, PCR ampli cation acetonitrile, and SpeedVac dried. The gel slices were rehydrated with was performed using forward primers directed against either an in- 12.5 ng/μl of mass spectrometry (MS)-grade Trypsin Gold (Promega, ternal region of algR (aglR-for; ATGAACGAAAGTGACGACATTTCC) or Madison, WI) in 40 mM NH4HCO3. The protease-generated peptides trpA (cccgaattcTTATGTGCGTTCCGGATGCG) together with a reverse were extracted with 0.1% (v/v) formic acid in 20 mM NH HCO , ′ 4 3 primer against a region downstream of aglR (aglR-5 downrev), re- followed by sonication for 20 min at room temperature, dehydration spectively yielding primer pairs a and b, or using primers aglR-for with 50% (v/v) acetonitrile, and additional sonication. After three rounds and aglR-rev (TCAACCAAGACTTTCAGATAGCAAC), designed to amplify a of extraction, the gel pieces were dehydrated with 100% acetonitrile, dried section of the aglR coding region (primer pair c). Reverse-transcription completely with a SpeedVac, resuspended in 5% (v/v) acetonitrile con- (RT)-PCR was performed as described previously [14], using primer pair taining 1% formic acid (v/v) and infused into the mass spectrometer fi ctotestforalgR transcription, so as to con rm aglR deletion at the RNA using static nanospray Econotips (New Objective, Woburn, MA). The level. protein digests were separated on-line by nano-flow reverse-phase liquid chromatography (LC) by loading onto a 150-mm by 75-μm (internal 2.3. Mass spectrometry diameter) by 365-μm (external diameter) Jupifer prepacked fused silica

5-μmC18 300 Å reverse-phase column (Thermo Fisher Scientific, Bremen, The total lipid contents of the Hfx. volcanii parent and ΔaglR cells Germany). The sample was eluted into the LTQ Orbitrap XL mass spec- were extracted and subjected to liquid chromatography-electrospray trometer (Thermo Fisher Scientific) using a 60-min linear gradient of ionization mass spectrometry (LC-ESI/MS) and tandem mass spec- 0.1% formic acid (v/v) in acetonitrile/0.1% formic acid (1:19, by volume) trometry (MS/MS) analysis as reported [10]. Cells were harvested to 0.1% formic acid in acetonitrile/0.1% formic acid (4:1, by volume) at a (8000 g, 30 min, 4 °C), resuspended in double-distilled water (DDW) flow rate of 300 nl/min. (1.33 ml DDW/g cells) and DNase (1.7 μg/ml; Sigma, St. Louis, MO) and stirred overnight at room temperature. Methanol and chloroform 2.4. [2-3H]-mannose radiolabeling were added to the cell extract to yield a methanol:chloroform:cell extract ratio of 2:1:0.8. After stirring for 24 h at room temperature, the mixture [2-3H]-mannose radiolabeling was performed according to [17]. was centrifuged (1075 g,30min,4°C).Thesupernatantfractionswere Cells of the parent and ΔaglR strains grown to mid-exponential collected, combined and filtered through glass wool. Chloroform and phase were incubated with 6 μl [2-3H]-mannose (23.8 mCi/mmol; DDW were added to the filtrate to yield a chloroform: DDW:filtrate PerkinElmer, Boston MA) in a volume of 100 μl. Sixty min later, the ratio of 1:1:3.8, in a separating funnel. After separation, the lower clear or- samples were precipitated with 15% (w/v) trichloroacetic acid and ganic phase, containing the total lipid extract, was collected into a round‐ separated by SDS-PAGE. The S-layer glycoprotein was identified by bottomed flask and evaporated in a rotary evaporator at 35 °C. For Coomassie-staining and fluorography and exposure to film. 1666 L. Kaminski et al. / Biochimica et Biophysica Acta 1820 (2012) 1664–1670

3. Results AglB, and was previously shown to occur upon deletion of other Hfx. volcanii N-glycosylation pathway components [9,10,14,18]. Although the 3.1. AglR contributes to Hfx. volcanii N-glycosylation unusual migration of the S-layer glycoprotein in SDS-PAGE (due to the reduced ability of the protein to bind SDS because of the highly acidic nature Transcription of a given open reading frame offers support for the as- of the protein [19])doesnotallowanyspecific conclusions as to the pre- signment of that sequence as corresponding to a true protein-encoding cise role of AglR, these results nonetheless reveal that AglR contributes to gene. Accordingly, previous RT-PCR efforts had revealed the transcription N-glycosylation of the Hfx. volcanii S-layer glycoprotein. of aglR [12]. Moreover, the co-transcription of aglR and aglE implies that AglR serves a role in Hfx. volcanii N-glycosylation. Now, as a first step in 3.2. In Hfx. volcanii cells lacking AglR, glycan-charged DolP accumulates directly determining whether AglR participates in N-glycosylation, Hfx. volcanii cells deleted of the encoding gene were generated according to Earlier findings reported that the first four residues of the the so-called ‘pop-in/pop-out’ technique developed for this organism pentasaccharide ultimately N-linked to the S-layer glycoprotein [15]. In this approach, the gene of interest (in this case, aglR)isreplaced are sequentially added to a common DolP carrier, whereas the inthegenomebytrpA, encoding tryptophan synthase, in a Hfx. volcanii final pentasaccharide residue (i.e. mannose) is derived from a dis- strain auxotrophic for tryptophan. To confirm replacement of aglR by tinct monosaccharide-charged DolP [7]. Thus, toward more precisely trpA at the DNA level, PCR was performed using primers directed at either defining the role of AglR, the glycan-charged DolP pool of ΔaglR cells aglR (primer pairs a and c) or trpA (primer pair b) (Fig. 1A). Whereas aglR was examined. Normal phase LC-ESI/MS [10] of glycan-charged C55 was solely detected in the parent strain, only the deletion strain contained and C60 DolP in parent and ΔaglR strain cells revealed that the level of trpA.Toconfirm deletion of aglR at the RNA level, RT-PCR was performed. tetrasaccharide-charged C55 DolP (m/z 776.449) was increased some Whereas a PCR product corresponding to aglR was generated when geno- 20-fold in ΔaglR cells, as compared to the parent strain, while mic DNA or cDNA prepared from parent strain cells served as template, no tetrasaccharide-charged C60 DolP (m/z 810.481) levels showed a close such products were generated when the same templates generated from to 13-fold increase (Fig. 2A; all peaks correspond to doubly-charged ΔaglR cells were used (not shown). [M−2H]2− ion species, unless otherwise noted). A similar phenome- In earlier studies on Hfx. volcanii N-glycosylation, efforts focused non was observed when the levels of the trisaccharide-charged lipid on the processing of the S-layer glycoprotein, a reporter of this carriers were considered in parent strain and mutant cells. In this post-translational modification in this species (for review, see [4]). case, the levels of trisaccharide-charged C55 and C60 DolP (m/z 681.419 The effects of aglR deletion on the S-layer glycoprotein were thus con- and 731.644, respectively) were increased 9- and 6-fold in ΔaglR cells, sidered by addressing the SDS-PAGE migration of this reporter in cells respectively (Fig. 2B). These increases are evident when one considers lacking AglR. In ΔaglR cells, the S-layer glycoprotein migrated faster than the unchanged levels of the singly-charged, non-DolP-related peaks at did the same protein from the parent strain (Fig. 1B). Such enhanced mi- m/z 805.685 and 731.644, relative to tetra- and trisaccharide-charged gration was also noted with cells lacking the oligosaccharyltransferase, C55 and C60 DolP in each strain. At the same time, no changes in the levels of the major Hfx. volcanii sulphated glycolipid, 6-HSO3-D-Manp- α1,2-D-Glcpα1,1-[sn-2,3-di-O-phytanylglycerol] (S-GL-1), present in the same sample injection as the glycan-charged DolP populations considered above were seen in the deletion strain cells ([M−H]− ion peak at m/z 1055.755 [20]), relative to the parent strain cells ([M−H]− ion peak at m/z 1055.737) (Fig. 2C). In the case of disaccharide-charged

C55 and C60 DolP, comparable levels were observed in the mutant and parent strain cells (not shown). When levels of monosaccharide-modified DolP species were considered, effects of aglR deletion were also observed. Hfx. volcanii contains several different monosaccharide-modified DolP species. The major species, generated through the actions of AglJ, serves as the hexose-charged DolP core onto which sugar residues two through four of the pentasaccharide ultimately N-linked to the S-layer glycoprotein are added [10]. A second, AglD-generated mannose-modified DolP species serves as the donor of the final N-linked pentasaccharide residue [7]. The predicted glycosyltransferase, HVO_1613, modifies a third hexose-bearing DolP [10], although the contribution of this species to N-glycosylation is unclear. Analysis of normal phase LC-extracted ion chromatograms (EIC) derived from the various hexose-modified DolP [M−H]− ions detected at m/z 1079.814 from parent and ΔaglR strain cells revealed the AglD-generated DolP-Man species (retained at 17.06 min in the parent strain) to be increased some 2.5-fold in cells Fig. 1. The absence of AglR affects Hfx. volcanii N-glycosylation. A. Left and middle panels: PCR lacking AglR (retained at 16.57 min in the ΔaglR strain) (Fig. 2D). In amplification was performed using a forward primer directed to a sequence within the aglR coding region and a reverse primer directed at the aglR 3′ flanking region (primer pair a) or contrast, aglR deletion had no significant effect on the levels of the using a forward primer directed to a sequence at the start of the trpA sequence and the same AglJ-generated species, with only 16% less of this species being detected reverse primer as above (primer pair b), together with genomic DNA from cells of the parent in that strain lacking AglR (retained at 15.81 min) than in the parent Δ strain (parent; left panel) or from cells where aglR had been replaced with trpA ( aglR; strain (retained at 16.30 min). Hence, in the absence of AglR, both middle panel), as template. Right panel: PCR amplification was performed using primer pair c directed against the aglR coding region, together with genomic DNA from cells of the tetrasaccharide- and mannose-linked dolichol phosphate accumulate. parent strain (parent) or the aglR-deleted strain (ΔaglR) as template. The positions to which the various primer pairs bind are shown in the drawing below the panels. Note that 3.3. In cells lacking AglR, the final mannose residue of the N-linked glycan aglR and trpA are surrounded by the same flanking regions. B. In the absence of AglR, the decorating the S-layer glycoprotein is not added Hfx. volcanii S-layer glycoprotein migrates faster in SDS-PAGE than does the same protein from the parent strain. Similarly enhanced migration is seen for the S-layer glycoprotein fi from cells lacking AglB, and hence incapable of performing N-glycosylation. Only that region To de ne further the contribution of AglR to N-glycosylation, parent of the Coomassie-stained gel containing the S-layer glycoprotein is shown. and ΔaglR strain S-layer glycoprotein-derived tryptic peptides, including L. Kaminski et al. / Biochimica et Biophysica Acta 1820 (2012) 1664–1670 1667

Fig. 2. Glycan-charged DolP accumulates in ΔaglR cells. Normal phase LC/MS/MS analysis of (A) tetra- and (B) trisaccharide-charged DolP from the total lipid extract of cells of the Hfx. volcanii parent strain (upper panels) and of ΔaglR cells (lower panels) was performed. A. Doubly charged [M−2H]2− ions of methyl ester of hexuronic 2− acid-dihexuronic acid-hexose-modified C55 and C60 DolPareshown.B.Doublycharged[M−2H] ions of dihexuronic acid-hexose-modified C55 and C60 DolP are shown. Non-DolP-related peaks at m/z 805.685 (A) and 731.644 (B) serve as internal controls for changes in glycan-charged DolP peaks as a function of aglR deletion. Note that different y-axis scales are used in the profiles of the parent and the ΔaglR cells. C. In both parent and the ΔaglR cells, identical levels of the sulfo-glycolipid, SGL-1 [20],aredetected as [M−H]− ion peaks at m/z 1055.7. D. EICs of the hexose-charged DolP [M−H]− ion at m/z 1079.8 from parent (upper panel) and ΔaglR strain cells (lower panel) are shown. The enzymes responsible for generating the three monosaccharide-charging DolP species are indicated above each peak. In ΔaglR cells, the AglD-generated species accumulates.

the N-terminal 1ERGNLDADSESFNK14 fragment that contains the glyco- To confirm both the earlier determination of mannose as the final sylated Asn-13 residue [5], were analyzed by LC-ESI/MS/MS [16].As residue of the N-linked pentasaccharide [7,11] and the importance of presented in Fig. 3 (left panels), the Asn-13-containing S-layer AglR for its addition, parent, ΔaglD and ΔaglR strain cells were glycoprotein-derived peptide generated from the parent strain is modi- incubated with 2-[3H]-mannose, and the incorporation of this fied by a pentasaccharide (doubly-charged [M−2H]2− ion peaks radiolabeled sugar into the S-layer glycoprotein was addressed. observed at m/z 1224.98, top left panel) comprising a hexose, two While parent strain cells readily incorporated 2-[3H]-mannose, no hexuronic acids, a methyl ester of hexuronic acid and a terminal such labeling was seen in cells deleted of aglD, encoding the mannose residue [5,6], as well as by precursor tetra- (m/z 1143.95), glycosyltransferase responsible for charging DolP with this final tri- (m/z 1048.42), di- (m/z 960.41) and monosaccharides (m/z 872.38) pentasaccharide residue [5] (Fig. 4, top panel, center and left lanes, (second through bottom left panels, respectively; in each case, the respectively). Likewise, when ΔaglR cells were challenged with doubly-charged [M−2H]2− ion peaks are shown). In cells lacking 2-[3H]-mannose, no radiolabel was incorporated into the S-layer gly- AglR,however,nopentasaccharide-modified Asn-13-containing peptide coprotein (Fig. 4, top panel, right lane). Densitometric quantification is observed (Fig. 3, top right panel). Yet, as observed in the parent strain, of the Coomassie-stained S-layer glycoprotein bands (Fig. 4, bottom versionsofthesamepeptidemodified by precursor tetra- (m/z 1143.95), panel) confirmed that comparable levels of S-layer glycoprotein tri- (m/z 1048.42), di- (m/z 960.41) and monosaccharides (m/z 872.39) from each strain had been loaded onto the gel (parent: 97.48±12.1, were detected (Fig. 3,secondtofifthpanelsontheright;ineachcase, ΔaglR: 87.14±12.1, ΔaglD: 91.44±6.6, (in each case, n=3); values the doubly-charged [M−2H]2− ion peaks are shown). in arbitrary units). 1668 L. Kaminski et al. / Biochimica et Biophysica Acta 1820 (2012) 1664–1670

Fig. 3. LC-ESI/MS/MS of an Asn-13-containing Hfx. volcanii S-layer glycoprotein-derived glycopeptide. The LC-ESI/MS/MS spectra of Asn-13-containing tryptic peptides derived from the S-layer glycoprotein from parent (left panels) or ΔaglR (right panels) strain cells are shown. The top to bottom panels show peaks corresponding to penta-, tetra-, tri-, di- and monosaccharide-modiﬁed peptides, as indicated. Above each peak, the doubly-charged [M−2H]2− ion species mass is indicated. L. Kaminski et al. / Biochimica et Biophysica Acta 1820 (2012) 1664–1670 1669

In the present study, it was shown that in Hfx. volcanii cells lacking AglR, there is an accumulation of DolP-Man and an absence of the final mannose residue from the N-linked glycan decorating the S-layer glycoprotein. Such observations are consistent with AglR acting as a DolP-Man flippase or contributing to DolP-Man flippase-related activity. Indeed, the homology of AglR to Wzx, a bacterial protein thought to translocate lipid-linked O-antigen precursor oligosaccharides across the plasma membrane [21–23], supports such a role for the Hfx. volcanii protein. Al- though it remains to be demonstrated that the accumulated DolP-Man seen in ΔaglR is restricted to the cytosolic face of the membrane, man- Fig. 4. ΔaglR cells lack the final mannose residue of the S-layer glycoprotein N-linked glycan. The S-layer glycoprotein of 2-[3H]-mannose-treated parent, ΔaglD and ΔaglR strain cells was nose is added to the protein-bound tetrasaccharide on the external face separated by SDS-PAGE and examined by fluorography (top panel; 2-[3H]-mannose) or of the membrane in the current working model of Hfx. volcanii Coomassie brilliant blue staining (bottom panel; CBB). N-glycosylation [11]. Indeed, the N-glycosylation of cell-impermeant peptides by a related haloarchaeal species, Halobacterium salinarum, points to the oligosaccharyltransferase, AglB as acting on the external surface of the cell [26]. At the same time, the accumulation of tetra- 3.4. AglR is a homologue of Wzx, a predicted lipid-linked oligosaccharide and trisaccharide-charged DolP species in the ΔaglR strain points to flippase AglR as serving alternate roles. For instance, AglR could serve a more general flippase-related function, such as regulating access to a flippase or Toward more precisely defining the role of AglR in Hfx. volcanii otherwise modulating flippase function. On the other hand, AglR could N-glycosylation, a bioinformatics approach was taken. The topology be required for the utilization but not the flipping of DolP-Man, as prediction servers, HMMTOP (http://www.enzim.hu/hmmtop/), SOSUI proposed for the eukaryal Lec35 protein [27]. It is also conceivable (http://bp.nuap.nagoya-u.ac.jp/sosui/), TMHMM (http://www.cbs.dtu. that AglR is responsible or otherwise involved in the transfer of dk/services/TMHMM-2.0/), TopPred (http://bioweb.pasteur.fr/seqanal/ the mannose residue from DolP-Man to the apparently target interfaces/toppred.html) and TMpred (http://www.ch.embnet.org/ protein-bound tetrasaccharide. This possibility is, however, unlike- software/TMPRED_form.html), all agreed that the 476 amino acid ly, as AglR does not contain any glycosyltransferase domains, such residue-containing AglR is a multi-membrane-spanning protein, as those found in the Hfx. volcanii glycosyltransferases, AglJ, AglG, containing 11 or 12 trans-membrane domains. An InterProScan search AglI, AglE and AglD. Moreover, unlike these glycosyltransferases, (http://www.ebi.ac.uk/Tools/InterProScan/) recognized a Polysacc_synt which present major soluble domains, AglR is predicted as being (PF01943) domain in AglR in the region between amino acids 155–204. largely buried within the membrane bilayer, spanning the mem- The Polysacc_synt domain is seen in protein involved in polysaccharide brane 11–12 times. Finally, one could envisage AglR being involved biogenesis and is found in RfbX (COG2244), also known as WzxB, a hy- in the membrane organization of other Agl proteins involved in Hfx. drophobic protein containing 12 potential trans-membrane domains volcanii N-glycosylation. [21,22]. The Wzx proteins are annotated as flippases responsible for de- If, however, AglR is indeed a DolP-Man flippase, then the fact that livering lipid-linked O-antigen precursor oligosaccharides across the monosaccharide-modified S-layer glycoprotein was detected in ΔaglR bacterial plasma membrane in an ATP-independent manner during li- cells suggests that AglR is able to distinguish between DolP species popolysaccharide biogenesis [21,23,24]. In the case of AglR, the various bearing mannose and other hexoses, in turn pointing to the involve- topology prediction algorithms assign the Polysacc_synt domain to a ment of multiple flippases in Hfx. volcanii N-glycosylation. A require- segment of the protein spanning the fifth and sixth trans-membrane ment for multiple flippases is also thought to be the case in eukaryal domains and partially facing the cell exterior. N-glycosylation [28,29]. Recently, the activity of a eukaryal DolP-Man To further examine whether AglR shares resemblance to Wzx pro- flippase was assayed in vitro [29]. When the same assay was attempted teins, several bacterial Wzx sequences were used as query in BLAST with Hfx. volcanii, namely assessing the ability of carboxy-2,2,6,6- searches against the Hfx. volcanii genome. Using the Escherichia coli tetramethylpiperidine 1-oxyl NO(+) to label externally exposed man- O32 Wzx sequence (ACD37057.1) as query, AglR was identified as a nose subunits, no labeling was detected, possibly to due to low levels homologue, with an e value of 2e−9. When this search was repeated of DolP-Man or effects related to the hypersaline conditions required using the E. coli O59 Wzx (AAV74382.1) or the E. coli O177 Wzx se- by Hfx. volcanii (not shown). Moreover, since the eukaryal DolP-Man quences (AAY728255.1) as query, AglR was again identified as a ho- flippase remains to be identified, sequence comparison with Hfx. mologue, with e values of 4e−5 and 1e−5, respectively. volcanii AglR is not yet possible. Likewise, a BLAST search of the eukaryal protein database failed to identify any AglR homologue. This may be re- 4. Discussion lated to the unique composition of DolP in Hfx. volcanii, relative to its eukaryal counterpart [7,30]. Based on the physico-chemical properties of phospholipids, cellu- At present, little is also known of LLO flippases in the other do- lar membranes present a hydrophobic barrier to the transfer of hy- mains of life, namely Eukarya and Bacteria. Based on genetic stud- drophilic molecules. Many biological processes, however, require ies, Rft1 was originally proposed as mediating the delivery of that this barrier be overcome. To achieve this, membranes contain a mannose5-N-acetylglucosamine2-charged dolichol pyrophosphate wide array of substrate-translocating proteins, including flippases, a (DolPP-M5) across the ER membrane in an ATP-independent man- class of proteins dedicated to the delivery of lipid species across ner [31,32]. Subsequent biochemical analysis of DolPP-M5 translo- membranes [25]. In N-glycosylation, flippases catalyze the transfer cation across ER-derived or proteoliposome membranes, however, of lipid-linked oligosaccharides (LLOs) from that side of the mem- revealed that Rft1 is not central to such activity [28,32,33].As brane where they are charged to the opposing membrane face, such, the agent responsible for DolPP-M5 flipping has yet to be where the LLO glycan cargo is delivered to other lipid-bound glycans defined and validated by genetic criteria. Indeed, it has been or to target proteins. Presently, relatively little is known of the flippases suggested that in the ER, Rft1 controls access to, rather than itself that catalyze the translocation of LLOs involved in N-glycosylation being, a flippase [28,34].Inthebacterium,Campylobacter jejuni, across membranes or their mechanisms of action. This is particularly PglK has been given the role of the ATP-dependent flippase of true in Archaea, where research efforts are hampered, in part, by a lack the N-glycosylation pathway, serving to translocate undecaprenol of appropriate molecular tools. pyrophosphate-linked heptasaccharide across the plasma membrane 1670 L. Kaminski et al. / Biochimica et Biophysica Acta 1820 (2012) 1664–1670

[35,36]. Nonetheless, some N-glycosylation persists in a pglK mutant, [15] T. Allers, H.P. Ngo, M. Mevarech, R.G. Lloyd, Development of additional selectable markers for the halophilic archaeon Haloferax volcanii based on the leuB and trpA thus calling this assignment into question [37]. genes, Appl. Environ. Microbiol. 70 (2004) 943–953. Defining the precise function of AglR will require in vitro reconsti- [16] D. Calo, Y. Eilam, R.G. Lichtenstein, J. Eichler, Towards glyco-engineering in tution of its activity. As the list of experimental tools available for Archaea: replacing Haloferax volcanii AglD with homologous glycosyltransferases from other halophilic archaea, Appl. Environ. Microbiol. 76 (2010) 5684–5692. working with Archaea, in general, and Hfx. volcanii, in particular, con- [17] D. Calo, Z. Guan, J. Eichler, Glyco-engineering in Archaea: differential N-glycosylation tinues to grow, such experiments may soon be possible. of the S-layer glycoprotein in a transformed Haloferax volcanii strain, Microb. Biotechnol. 4 (2011) 461–470. [18] S. Yurist-Doutsch, H. Magidovich, V.V. Ventura, P.G. Hitchen, A. Dell, J. Eichler, N-glycosylation in Archaea: on the coordinated actions of Haloferax volcanii Acknowledgements AglF and AglM, Mol. Microbiol. 75 (2010) 1047–1058. [19] M. Sumper, E. Berg, R. Mengele, I. Strobel, Primary structure and glycosylation of JE is supported by grants from the Israel Science Foundation the S-layer protein of Haloferax volcanii, J. Bacteriol. 172 (1990) 7111–7118. fi [20] S. Naparstek, Z. Guan, J. Eichler, The predicted geranylgeranyl reductase, (30/07) and the US Army Research Of ce (W911NF-11-1-520). HVO_1799, reduces the omega position isoprene of dolichol phosphate in the The mass spectrometry facility in the Department of Biochemistry halophilic archaeon, Haloferax volcanii, Biochim. Biophys. Acta, Mol. Cell Biol. of the Duke University Medical Center and ZG are supported by the Lipids 1821 (2012) 923–933. LIPID MAPS Large Scale Collaborative Grant (GM-069338) from the [21] D. Liu, R.A. Cole, P.R. Reeves, An O-antigen processing function for Wzx (RfbX), a promising candidate for O-unit flippase, J. Bacteriol. 178 (1996) 2102–2107. NationalInstitutes of Health. LK is the recipient of a Negev-Zin As- [22] Z. Yao, M.A. Valvano, Genetic analysis of the O-specific lipopolysaccharide biosynthesis sociates Scholarship. region (rfb) of Escherichia coli K-12 W3110: identification of genes that confer group 6 specificity to Shigella flexneri serotypes Y and 4a, J. Bacteriol. 176 (1994) 4133–4143. [23] C.R.H. Raetz, C. Whitfield, Lipopolysaccharide endotoxins, Annu. Rev. Biochem. 71 (2002) 635–700. References [24] P.D. Rick, K. Barr, K. Sankaran, J. Kajimura, J.S. Rush, C.J. Waechter, Evidence that the wzxE gene of Escherichia coli K-12 encodes a protein involved in the trans- [1] A. Helenius, M. Aebi, Roles of N-linked glycans in the endoplasmic reticulum, bilayer movement of a trisaccharide-lipid intermediate in the assembly of entero- Annu. Rev. Biochem. 73 (2004) 1019–1049. bacterial common antigen, J. Biol. Chem. 278 (2003) 16534–16542. [2] C.M. Szymanski, B.W. Wren, Protein glycosylation in bacterial mucosal pathogens, [25] S. Sanyal, A.K. Menon, Flipping lipids: why an' what's the reason for? ACS Chem. Nat. Rev. Microbiol. 3 (2005) 225–237. Biol. 4 (2009) 895–909. [3] E. Weerapana, B. Imperiali, Asparagine-linked protein glycosylation: from eukaryotic [26] J. Lechner, F. Wieland, M. Sumper, Transient methylation of dolichyl oligosaccharides to prokaryotic systems, Glycobiology 16 (2006) 91R–101R. is an obligatory step in halobacterial sulfated glycoprotein biosynthesis, J. Biol. Chem. [4] D. Calo, L. Kaminski, J. Eichler, Protein glycosylation in Archaea: sweet and ex- 260 (1985) 8984–8989. treme, Glycobiology 20 (2010) 1065–1079. [27] M. Anand, J.S. Rush, S. Ray, M.A. Doucey, J. Weik, F.E. Ware, J. Hofsteenge, C.J. [5] M. Abu-Qarn, S. Yurist-Doutsch, A. Giordano, A. Trauner, H.R. Morris, P. Hitchen, A. Waechter, M.A. Lehrman, Requirement of the Lec35 gene for all known classes Dell, J. Eichler, Haloferax volcanii AglB and AglD are involved in N-glycosylation of of monosaccharide-P-dolichol-dependent glycosyltransferase reactions in mam- the S-layer glycoprotein and proper assembly of the surface layer, J. Mol. Biol. 14 mals, Mol. Biol. Cell 12 (2001) 487–501. (2007) 1224–1236. [28] S. Sanyal, C.G. Frank, A.K. Menon, Distinct flippases translocate glycerophospholipids [6] H.Magidovich,S.Yurist-Doutsch,Z.Konrad,V.V.Ventura,A.Dell,P.G.Hitchen,J.Eichler, and oligosaccharide diphosphate dolichols across the endoplasmic reticulum, AglP is a S-adenosyl-L-methionine-dependent methyltransferase that participates in Biochemistry 47 (2008) 7937–7946. the N-glycosylation pathway of Haloferax volcanii,Mol.Microbiol.76(2010) [29] S. Sanyal, A.K. Menon, Stereoselective transbilayer translocation of mannosyl 190–199. phosphoryl dolichol by an endoplasmic reticulum flippase, Proc. Natl. Acad. Sci. [7] Z. Guan, S. Naparstek, L. Kaminski, Z. Konrad, J. Eichler, Distinct glycan-charged phos- U. S. A. 107 (2010) 11289–11294. phodolichol carriers are required for the assembly of the pentasaccharide N-linked [30] C. Kuntz, J. Sonnenbichler, I. Sonnenbichler, M. Sumper, R. Zeitler, Isolation and charac- to the Haloferax volcanii S-layer glycoprotein, Mol. Microbiol. 78 (2010) 1294–1303. terization of dolichol-linked oligosaccharides from Haloferax volcanii, Glycobiology 7 [8]M.Abu-Qarn,A.Giordano,F.Battaglia,A.Trauner,P.Hitchen,H.R.Morris,A.Dell,J. (1997) 897–904. Eichler, Identification of AglE, a second glycosyltransferase involved in N glycosylation [31] D.T. Ng, E.D. Spear, P. Walter, The unfolded protein response regulates multiple of the Haloferax volcanii S-layer glycoprotein, J. Bacteriol. 190 (2008) 3140–3146. aspects of secretory and membrane protein biogenesis and endoplasmic reticu- [9] S. Yurist-Doutsch, M. Abu-Qarn, F. Battaglia, H.R. Morris, P.G. Hitchen, A. Dell, J. lum quality control, J. Cell Biol. 150 (2000) 77–88. Eichler, AglF, aglG and aglI, novel members of a gene island involved in the [32] J. Helenius, D.T. Ng, C.L. Marolda, P. Walter, M.A. Valvano, M. Aebi, Translocation N-glycosylation of the Haloferax volcanii S-layer glycoprotein, Mol. Microbiol. 69 of lipid-linked oligosaccharides across the ER membrane requires Rft1 protein, (2008) 1234–1245. Nature 415 (2002) 447–450. [10] L. Kaminski, M. Abu-Qarn, Z. Guan, S. Naparstek, V.V. Ventura, C.R.H. Raetz, P.G. Hitchen, [33] C.G. Frank, S. Sanyal, J.S. Rush, C.J. Waechter, A.K. Menon, Does Rft1 flip an A. Dell, J. Eichler, AglJ adds the first sugar of the N-linked pentasaccharide decorating N-glycan lipid precursor? Nature 454 (2008) E3–E4. the Haloferax volcanii S-layer glycoprotein, J. Bacteriol. 192 (2010) 5572–5579. [34] J.S. Rush, N. Gao, M.A. Lehrman, S. Matveev, C.J. Waechter, Suppression of Rft1 ex-

[11] D. Calo, Z. Guan, S. Naparstek, J. Eichler, Different routes to the same ending: compar- pression does not impair the transbilayer movement of Man5GlcNAc2-P-P-dolichol ing the N-glycosylation processes of Haloferax volcanii and Haloarcula marismortui, in sealed microsomes from yeast, J. Biol. Chem. 284 (2009) 19835–19842. two halophilic archaea from the Dead Sea, Mol. Microbiol. 81 (2011) 1166–1177. [35] D. Linton, N. Dorrell, P.G. Hitchen, S. Amber, A.V. Karlyshev, H.R. Morris, A. Dell, [12] S. Yurist-Doutsch, J. Eichler, Manual annotation, transcriptional analysis and protein ex- M.A. Valvano, M. Aebi, B.W. Wren, Functional analysis of the Campylobacter jejuni pression studies reveal novel genes in the agl cluster responsible for N-glycosylation in N-linked protein glycosylation pathway, Mol. Microbiol. 55 (2005) 1695–1703. the halophilic archaeon Haloferax volcanii, J. Bacteriol. 191 (2009) 3068–3075. [36] J. Kelly, H. Jarrell, L. Millar, L. Tessier, L.M. Fiori, P.C. Lau, B. Allan, C.M. Szymanski, [13] M. Mevarech, R. Werczberger, Genetic transfer in Halobacterium volcanii,J.Bacteriol. Biosynthesis of the N-linked glycan in Campylobacter jejuni and addition onto 162 (1985) 461–462. protein through block transfer, J. Bacteriol. 188 (2006) 2427–2434. [14] M. Abu-Qarn, J. Eichler, Protein N-glycosylation in Archaea: deﬁning Haloferax [37] C. Alaimo, I. Catrein, L. Morf, C.L. Marolda, N. Callewaert, M.A. Valvano, M.F. volcanii genes involved in S-layer glycoprotein glycosylation, Mol. Microbiol. 61 Feldman, M. Aebi, Two distinct but interchangeable mechanisms for ﬂipping of (2006) 511–525. lipid-linked oligosaccharides, EMBO J. 25 (2006) 967–976.

Chapter 3.5

"Two distinct N-glycosylation pathways together process the S-layer

glycoprotein in the halophilic archaea,

Haloferax volcanii"

Lina Kaminski, Ziqiang Guan, Sophie Yurist-Doutsch and Jerry Eichler mBio. 2013a. 4:e00716-13

Abstract:

N-glycosylation in Archaea presents aspects of this post-translational modification not seen in either Eukarya or Bacteria. In the haloarchaeon Haloferax volcanii, the surface (S)-layer glycoprotein can be simultaneously modified by two different N-glycans. Asn-13 and Asn-83 are modified by a pentasaccharide, whereas

Asn-498 is modified by a tetrasaccharide of distinct composition, with N- glycosylation at this position being related to environmental conditions. Specifically,

N-glycosylation of Asn-498 is detected when cells were grown in the presence of 1.75 but not 3.4 M NaCl. While deletion of genes encoding components of the pentasaccharide assembly pathway had no effect on the biosynthesis of the tetrasaccharide bound to Asn-498, deletion of genes within the cluster spanning

HVO_2046-HVO_2061 interfered with the assembly and attachment of the Asn-498- linked tetrasaccharide. Transfer of the ‘low salt’ tetrasaccharide from the dolichol phosphate carrier upon which it is assembled to S-layer glycoprotein Asn-498 did not require AglB, the oligosaccharyltransferase responsible for pentasaccharide attachment to Asn-13 and Asn-83. Finally, although biogenesis of the ‘low salt’ tetrasaccharide is barely discernible upon growth at the elevated salinity, this glycan was readily detected under such conditions in strains deleted of pentasaccharide

biosynthesis pathway genes, indicative of crosstalk between the two N-glycosylation pathways.

Importance: In the haloarchaea Haloferax volcanii, originally from the Dead

Sea, the pathway responsible for the assembly and attachment of a pentasaccharide to the S-layer glycoprotein, a well-studied glycoprotein in this species, has been described. More recently, it was shown that in response to growth in low salinity, the same glycoprotein is modified by a novel tetrasaccharide. In the present study, numerous components of the pathway used to synthesize this ‘low salt’ tetrasaccharide are described. As such, this represents the first report of two N- glycosylation pathways able to simultaneously modify a single protein as a function of environmental salinity. Moreover, and to the best of our knowledge, the ability to

N-glycosylate the same protein with different and unrelated glycans has not been observed in either Eukarya or Bacteria, or indeed beyond the halophilic archaea, where similar dual modification of the Halobacterium salinarum S-layer glycoprotein was reported.

Downloaded from

Two Distinct N-Glycosylation Pathways mbio.asm.org Process the Haloferax volcanii S-Layer Glycoprotein upon Changes in

Environmental Salinity on November 18, 2013 - Published by

Lina Kaminski, Ziqiang Guan, Sophie Yurist-Doutsch, et al. 2013. Two Distinct N-Glycosylation Pathways Process the Haloferax volcanii S-Layer Glycoprotein upon Changes in Environmental Salinity mBio. 4(6): . doi:10.1128/mBio.00716-13.

Updated information and services can be found at: http://mbio.asm.org/content/4/6/e00716-13.full.html mbio.asm.org

SUPPLEMENTAL http://mbio.asm.org/content/4/6/e00716-13.full.html#SUPPLEMENTAL MATERIAL

REFERENCES This article cites 35 articles, 12 of which can be accessed free at: http://mbio.asm.org/content/4/6/e00716-13.full.html#ref-list-1 CONTENT ALERTS Receive: RSS Feeds, eTOCs, free email alerts (when new articles cite this article), more>>

Information about commercial reprint orders: http://mbio.asm.org/misc/reprints.xhtml Information about Print on Demand and other content delivery options: http://mbio.asm.org/misc/contentdelivery.xhtml To subscribe to another ASM Journal go to: http://journals.asm.org/subscriptions/

Downloaded from

RESEARCH ARTICLE

Two Distinct N-Glycosylation Pathways Process the Haloferax volcanii mbio.asm.org S-Layer Glycoprotein upon Changes in Environmental Salinity on November 18, 2013 - Published by Lina Kaminski,a Ziqiang Guan,b Sophie Yurist-Doutsch,a* Jerry Eichlera Department of Life Sciences, Ben Gurion University of the Negev, Beersheva, Israela; Department of Biochemistry, Duke University Medical Center, Durham, North Carolina, USAb * Present address: Sophie Yurist-Doutsch, Michael Smith Laboratories, University of British Columbia, Vancouver, BC, Canada.

ABSTRACT N-glycosylation in Archaea presents aspects of this posttranslational modification not seen in either Eukarya or Bac- teria. In the haloarchaeon Haloferax volcanii, the surface (S)-layer glycoprotein can be simultaneously modified by two different N-glycans. Asn-13 and Asn-83 are modified by a pentasaccharide, whereas Asn-498 is modified by a tetrasaccharide of distinct composition, with N-glycosylation at this position being related to environmental conditions. Specifically, N-glycosylation of Asn-498 is detected when cells are grown in the presence of 1.75 but not 3.4 M NaCl. While deletion of genes encoding components of the pentasaccharide assembly pathway had no effect on the biosynthesis of the tetrasaccharide bound to Asn-498, deletion of genes within the cluster spanning HVO_2046 to HVO_2061 interfered with the assembly and attachment of the Asn-498- mbio.asm.org linked tetrasaccharide. Transfer of the “low-salt” tetrasaccharide from the dolichol phosphate carrier upon which it is assembled to S-layer glycoprotein Asn-498 did not require AglB, the oligosaccharyltransferase responsible for pentasaccharide attachment to Asn-13 and Asn-83. Finally, although biogenesis of the low-salt tetrasaccharide is barely discernible upon growth at the elevated salinity, this glycan was readily detected under such conditions in strains deleted of pentasaccharide biosynthesis pathway genes, indicative of cross talk between the two N-glycosylation pathways. IMPORTANCE In the haloarchaeon Haloferax volcanii, originally from the Dead Sea, the pathway responsible for the assembly and attachment of a pentasaccharide to the S-layer glycoprotein, a well-studied glycoprotein in this species, has been described. More recently, it was shown that in response to growth in low salinity, the same glycoprotein is modified by a novel tetrasaccharide. In the present study, numerous components of the pathway used to synthesize this “low-salt” tetrasaccharide are described. As such, this represents the first report of two N-glycosylation pathways able to simultaneously modify a single protein as a function of environmental salinity. Moreover, and to the best of our knowledge, the ability to N-glycosylate the same protein with different and unrelated glycans has not been observed in either Eukarya or Bacteria or indeed beyond the halophilic archaea, for which similar dual modification of the Halobacterium salinarum S-layer glycoprotein was reported.

Received 28 August 2013 Accepted 10 October 2013 Published 5 November 2013 Citation Kaminski L, Guan Z, Yurist-Doutsch S, Eichler J. 2013. Two distinct N-glycosylation pathways process the Haloferax volcanii S-layer glycoprotein upon changes in environmental salinity. mBio 4(6):e00716-13. doi:10.1128/mBio.00716-13. Invited Editor Maria Hadjifrangiskou, Vanderbilt University School of Medicine Editor Scott Hultgren, Washington University School of Medicine Copyright © 2013 Kaminski et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-Noncommercial-ShareAlike 3.0 Unported license, which permits unrestricted noncommercial use, distribution, and reproduction in any medium, provided the original author and source are credited. Address correspondence to Jerry Eichler, [email protected].

-glycosylation, the covalent attachment of glycans to select AglJ, AglG, AglI, and AlgE sequentially add the first four sugars of Nasparagine residues of target proteins, is a posttranslational the pentasaccharide to a common dolichol phosphate (DolP) car- modification performed by members of all three domains of life rier (15–18). Once the lipid-linked tetrasaccharide (and its pre- (1–5). N-glycosylation in Archaea, although seemingly common cursors) is translocated across the plasma membrane, the glycan is (6) and involving a diversity of sugars not seen elsewhere (4, 5), N-linked to the S-layer glycoprotein by the oligosaccharyltrans- nonetheless remains less well understood than the parallel eu- ferase (OST) AglB (13). The final sugar of the N-linked pentasac- karyal and bacterial systems. Of late, however, a series of genetic charide, mannose, is added to its own DolP carrier on the cyto- and biochemical studies of Haloferax (Hfx.) volcanii (see reference plasmic face of the membrane by AglD, translocated to face the 5) and other species (7–12) have begun to provide insight into the cell exterior in a process involving AglR, and then transferred to archaeal version of this universal protein-processing event. the protein-linked tetrasaccharide by AglS (17, 19–22). Finally, In Hfx. volcanii, the surface (S)-layer glycoprotein, a well- other Agl proteins, such as AglF, AglM, and AglP, also contribute studied glycoprotein and sole component of the protein shell sur- to pentasaccharide assembly, serving various sugar-processing rounding cells of this species, is modified by a pentasaccharide roles (14–23). comprising a hexose, two hexuronic acids, a methyl ester of hexu- Hfx. volcanii, first isolated from the Dead Sea (24), requires ronic acid, and a mannose (13, 14) via the actions of a series of Agl molar concentrations of salt for survival. Recently, it was reported (archaeal glycosylation) proteins. Here, the glycosyltransferases that N-glycosylation of the S-layer glycoprotein, in direct contact

November/December 2013 Volume 4 Issue 6 e00716-13 ® mbio.asm.org 1 Downloaded from

Kaminski et al. with the hypersaline surroundings in which such cells exist, same low-salt tetrasaccharide was also detected when an S-layer 497 changes as a function of environmental salinity (25). S-layer gly- glycoprotein-derived Asn-498-containing peptide ( INGTAS mbio.asm.org coproteins Asn-13 and Asn-83 bear the pentasaccharide described GANSVLVIFVD513; calculated [M ϩ 2H]2ϩ mass, 1,675.88 Da) above in cells raised in either 1.75 M or 3.4 M NaCl-containing from cells lacking AglI grown in low-salt medium was examined growth medium, although such modification is substantially re- by LC-ESI MS. The resulting profile included an m/z 1,195.03 duced upon growth at the lower salinity. More strikingly, S-layer peak (Fig. 1A, bottom), as well as m/z 1,122.01, 1,040.97, and glycoprotein Asn-498, whose position is not modified in cells 959.95 peaks (see Fig. S1 in the supplemental material, left col- on November 18, 2013 - Published by grown at the higher salinity, is modified by a novel tetrasaccharide umn). These values are in good agreement with, respectively, the comprising a sulfated hexose, two hexoses, and a rhamnose in cells predicted [M ϩ 2H]2ϩ values of the Asn-498-containing peptide grown at the lower salt concentration. The modulation of the modified by the complete low-salt tetrasaccharide (m/z 1,194.94) S-layer glycoprotein N-glycosylation profile in response to chang- and precursors comprising the first three (m/z 1,121.94), the first ing environmental conditions represents a novel role for this com- two (m/z 1,040.94), and the first (m/z 959.94) sugar components mon posttranslational modification. of this glycan (25). Likewise, in keeping with the role of AglE in To the best of our knowledge, the ability to N-glycosylate the adding the fourth pentasaccharide sugar, a hexuronic acid that is same protein with different and unrelated glycans has not been subsequently methylated by AglP (14, 15), the LC-ESI MS profile observed in either Eukarya or Bacteria or indeed beyond the halo- of proteolytic S-layer glycoprotein fragments from cells lacking philic archaea, for which similar dual modification of the Halo- AglE did not present a peak corresponding to the Asn-83- bacterium (Hbt.) salinarum S-layer glycoprotein was previously containing peptide modified by the first four pentasaccharide sug- 2ϩ reported (26). Still, it is not known whether the Agl pathway, ars (m/z 1,674.675, calculated [M ϩ 2H] mass) (Fig. 1B, top, mbio.asm.org responsible for the assembly and attachment of the pentasac- left). Instead, an [M ϩ 2H]2ϩ peak at m/z 1,579.72, corresponding charide bound at the Hfx. volcanii S-layer glycoprotein Asn-13 to the same peptide modified by the first three sugars of the Asn- and Asn-83 positions, is also involved in generating the “low-salt” 83-bound pentasaccharide was seen (Fig. 1B, top, right). LC-ESI tetrasaccharide bound to Asn-498 or whether a novel N- MS analysis of a total lipid extract of the same cells revealed [M- 2Ϫ glycosylation pathway is responsible. In the present study, this was 2H] peaks at m/z 780.448 and 814.475, corresponding to C55 addressed. and C60 DolPs modified by the low-salt tetrasaccharide, respectively (Fig. 1B, middle). At the same time, the LC-ESI MS profile of RESULTS the Asn-498-containing peptide from such cells presented [M ϩ Assembly of the S-layer glycoprotein Asn-13- and Asn-83- 2H]2ϩ peaks at m/z 1,195.04 (Fig. 1B, bottom), 1,122.03, and linked pentasaccharide and the Asn-498-linked low-salt tetra- 1,040.97 (Fig. S1, right column), respectively, corresponding to saccharide rely on different pathways. To determine whether the the low-salt tetrasaccharide and its tri- and disaccharide precur- same Agl pathway responsible for generating the pentasaccharide sors. N-linked to S-layer glycoprotein Asn-13 and Asn-83 (13, 14) is These observations thus point to the assembly of the low-salt also responsible for generating the low-salt tetrasaccharide bound tetrasaccharide bound to S-layer glycoprotein Asn-498 as relying to Asn-498 (25), N-glycosylation at this position was considered on a biosynthetic pathway distinct from that used for assembly of in Hfx. volcanii cells lacking AglI or AglE, glycosyltransferases in- the pentasaccharide that modifies the Asn-13 and Asn-83 posi- volved in pentasaccharide assembly (15, 16), and grown in 1.75 M tions. NaCl-containing (low-salt) medium. These mutants were selected A novel N-glycosylation pathway modifies S-layer glycopro- for such analysis, as the shortened N-linked glycan decorating tein Asn-498. Apart from aglD, all of the Hfx. volcanii genes in- Asn-13 and Asn-83 in each case is readily detected by mass spec- volved in the assembly of the N-linked pentasaccharide decorating trometry. S-layer glycoprotein Asn-13 and Asn-83 are organized into a sin- To assess the contributions of AglI and AglE to Asn-498 glyco- gle cluster spanning HVO_1517 (aglJ)toHVO_1531 (aglM) (16, sylation, liquid chromatography-electrospray ionization mass 27). To begin identifying components of the novel pathway in- spectrometry (LC-ESI MS) was employed. Initially, the LC-ESI volved in assembly of the low-salt tetrasaccharide, the Hfx. vol- MS profile of an S-layer glycoprotein-derived Asn-83-containing canii genome sequence (28) was scanned for other clusters of open peptide generated by trypsin and Glu-C protease treatment reading frames (ORFs) annotated as serving glycosylation-related (65NQPLGTYDVDGSGSATTPNVTLLAPR90) from ⌬aglI cells roles. ORFs comprising the region spanning from HVO_2046 to grown in low-salt medium was considered. Whereas no peptide HVO_2061 represents one such region. Table S1 in the supple- modified by the first three pentasaccharide sugars (m/z 1,597.675, mental material summarizes the current annotations of these se- calculated [M ϩ 2H]2ϩ mass) was detected (Fig. 1A, top, left), a quences, as well as the predicted subcellular localization of the peptide modified by the first two pentasaccharide sugars, a hexose deduced products. and a hexuronic acid, was observed ([M ϩ 2H]2ϩ peak at m/z The participation of genes within the HVO_2046-2061 region 1,491.70) (Fig. 1A, top, right). These results reflect the previously in generating the low-salt tetrasaccharide bound to Asn-498 was reported role of AglI in adding the third pentasaccharide sugar, a next considered by deleting each sequence within the cluster and hexuronic acid, to disaccharide-charged DolP, from where it is assessing the effects of such deletion on Asn-498 glycosylation. transferred to S-layer glycoprotein Asn-83 (16). At the same time, The various deletion strains were generated by the “pop-in/pop- when a total lipid extract of the ⌬aglI cells grown in low-salt me- out” technique developed for use with Hfx. volcanii, in which the dium was examined, [M-2H]2Ϫ peaks at m/z 780.457 and 814.479, trpA gene, encoding tryptophan synthase, is introduced into a corresponding to C55 and C60 DolPs modified by a sulfated strain auxotrophic for tryptophan in place of the gene being de- hexose, two hexoses, and a rhamnose, namely, the complete low- leted (29). In each case, deletion was confirmed (Fig. S2) and the salt tetrasaccharide (25), were observed (Fig. 1A, middle). The effect of each deletion was assessed by LC-ESI MS both at the level

2 ® mbio.asm.org November/December 2013 Volume 4 Issue 6 e00716-13 Downloaded from

Two Hfx. volcanii N-Glycosylation Pathways

of the DolP carrier upon which the low-salt tetrasaccharide is

assembled (25) and at the level of S-layer glycoprotein Asn-498. mbio.asm.org In the case of ⌬HVO_2058 cells, analysis of the LC-ESI MS

profile of a total lipid extract revealed C55 and C60 DolPs modified by the first three sugars of the low-salt tetrasaccharide ([M-2H]2Ϫ ion peaks at m/z 707.415 and 741.443, respectively) (Fig. 2A, left) but not by the complete tetrasaccharide (Fig. 2A, right). Likewise, on November 18, 2013 - Published by when the Asn-498-containing S-layer glycoprotein-derived peptide from the same cells grown in low-salt medium was analyzed, a[Mϩ 2H]2ϩ ion peak at m/z 1,122.03 was observed, corresponding to the peptide modified by the trisaccharide precursor of the low-salt tetrasaccharide (Fig. 2B, upper). No peptide modified by the complete tetrasaccharide was detected (Fig. 2B, lower). HVO_2058 is thus involved in the addition of rhamnose onto DolP charged with the first three sugars of the low-salt tetrasaccharide. To confirm that the effect of HVO_2058 deletion on low- salt tetrasaccharide biogenesis was not due to a downstream effect, cells of the deletion strain were transformed to express a plasmid-

derived version of the protein bearing a Clostridium thermocellum mbio.asm.org cellulose-binding domain (CBD) tag. The transformed strain was able to assemble and attached the complete low-salt tetrasaccharide to S-layer glycoprotein Asn-498 (Fig. 2C). Finally, to gain insight into HVO_2058 transcription as a function of growth medium salinity, quantitative PCR was performed. Relative to the level of HVO_2058 transcript measured in cells grown in 3.4 M NaCl-containing medium, 0.75-fold (Ϯ0.22-fold, standard deviation, n ϭ 5) the transcript was seen in cells grown in 1.75 M NaCl-containing medium. As such, comparable levels of the HVO_2058 transcript are found at the different levels of salinity. ⌬ At the same time, C55 and C60 DolPs in HVO_2048 cells were modified by only the first two sugars of the low-salt tetrasaccharide ([M-2H]2Ϫ ion peaks at m/z 626.382 and 660.408) (Fig. S3A, left); no peaks corresponding to the same lipids bearing the first three sugars of the low-salt tetrasaccharide were detected (Fig. S3A, right). Analysis of the Asn-498-containing peptide from these cells revealed a [M ϩ 2H]2ϩ ion peak of m/z 1,040.97 (Fig. S3B, upper), corresponding to the peptide modified by the first two sugars of the low-salt tetrasaccharide; no peptide modified by the trisaccharide precursor of the Asn-498-linked low-salt tetrasaccharide was observed (Fig. S3B, lower). As such, it can be concluded that HVO_2048 is involved in the addition of the third sugar of the low-salt tetrasaccharide to DolP. To assess the effects of growth medium salinity on HVO_2048 transcription, quantitative PCR was performed. Relative to the level of HVO_2048 transcript measured in cells grown in 3.4 M NaCl-containing medium, 0.76-fold (Ϯ0.44-fold, standard deviation, n ϭ 5) of the transcript was seen in cells grown in 1.75 M NaCl-containing medium. It thus appears that the level of HVO_2048 transcription does not change substantially as a function of medium salinity. Using the same experimental strategy, roles for HVO_2046, HVO_2049, HVO_2053, HVO_2055, HVO_2056, HVO_2057,

Figure Legend Continued FIG 1 The Agl pathway is not involved in biogenesis of the low-salt tetrasaccharide. In ⌬aglI and ⌬aglE cells grown in low-salt medium, S-layer glyco- positions, while in the lipid-derived profiles, arrows point to monoisotopic 2Ϫ protein Asn-498 is modified by the low-salt tetrasaccharide. LC-ESI MS pro- [M-2H] ion peaks corresponding to low-salt tetrasaccharide-modified files of S-layer glycoprotein-derived peptides containing Asn-83 (top) or Asn- DolP. Adjacent to each arrow is a schematic depiction of the Asn- or DolP- 498 (bottom) or of total lipid extracts (middle) from ⌬aglI (A) and ⌬aglE (B) bound glycan detected (or not), with the filled circles corresponding to hexose, ϩ the filled squares corresponding to hexuronic acid, and the open circle corre- cells. In each peptide-derived profile, arrows point to monoisotopic [M ϫ 2H]2ϩ ion peaks corresponding to glycan-modified peptides or their expected sponding to rhamnose. In the middle panels, the arrows indicating 10 reflect magnification of the ion peaks in the corresponding region of the profile. (Continued)

November/December 2013 Volume 4 Issue 6 e00716-13 ® mbio.asm.org 3 Downloaded from

Kaminski et al. mbio.asm.org on November 18, 2013 - Published by mbio.asm.org

FIG 2 Deletion of HVO_2058 affects DolP and Asn-498 glycosylation. (A) In cells lacking HVO_2058, C55 and C60 DolPs are modified by the first three sugars of the complete low-salt tetrasaccharide (left) but not by the complete glycan (right). (B) The S-layer glycoprotein-derived Asn-498-containing peptide from ⌬HVO_2058 cells is modified by the first three sugars of the low-salt tetrasaccharide (upper) but not by the complete glycan (lower). (C) In cells of the deletion strain transformed with plasmid pWL-CBD2058, encoding a CBD-tagged version of HVO_2058, the S-layer glycoprotein-derived Asn-498-containing peptide is modified by the complete low-salt tetrasaccharide. The inset shows PCR amplification of CBD-HVO_2058 in the transformed strain but not in the deletion strain. In each panel in the figure, the position of each detected or absent glycan-modified DolP or Asn-498-containing peptide is indicated, as is a schematically drawn depiction of the present or absent glycan. In the schematic drawings, the open circle represents rhamnose, while the filled circles represent hexose.

HVO_2059, HVO_2060, and HVO_2061 in Hfx. volcanii S-layer ride, or it mediates an effect not revealed by the mass glycoprotein Asn-498 glycosylation were also revealed (Table 1). spectrometry-based analysis employed, as, for example, would be In the case of HVO_2047, deletion of the encoding gene did not the case for an epimerase. reveal any effect on the low-salt tetrasaccharide at either the DolP Given the roles of products of genes in the HVO_2046-2061 or the S-layer glycoprotein level. As such, either HVO_2047 is not region in the biogenesis of the low-salt tetrasaccharide, these involved in N-glycosylation, as related to the low-salt tetrasaccha- were renamed Agl5 (HVO_2053), Agl6 (HVO_2061), Agl7

4 ® mbio.asm.org November/December 2013 Volume 4 Issue 6 e00716-13 Downloaded from

Two Hfx. volcanii N-Glycosylation Pathways

TABLE 1 Genes involved in the biogenesis of the low-salt tetrasaccharidea mbio.asm.org Deletion (gene designation) C55 or C60-DolP species observed S-layer glycoprotein Asn-498 glycosylation ⌬ Œ HVO_2046 (agl7) DolP- N498- ⌬ HVO_2048 (agl9) DolP- SO3- N498- SO3- ⌬ HVO_2049 (agl10) DolP- SO3- N498- SO3- ⌬HVO_2053 (agl5) DolP- Not observed ⌬ Œ on November 18, 2013 - Published by HVO_2055 (agl15) DolP- SO3- Not observed ⌬ HVO_2056 (agl13) DolP- SO3- N498- SO3- ⌬ HVO_2057 (agl11) DolP- SO3- N498- SO3- ⌬ HVO_2058 (agl14) DolP- SO3- N498- SO3- ⌬ HVO_2059 (agl12) DolP- SO3- N498- SO3- ⌬ HVO_2060 (agl8) DolP- SO3- N498- SO3- ⌬HVO_2061 (agl6) DolP- Not observed a Filled circles represent hexose, and the open circles represent rhamnose.

(HVO_2046), Agl8 (HVO_2060), Agl9 (HVO_2048), Agl10 high-salt medium contained barely detectable amounts of C55 and 2Ϫ (HVO_2049), Agl11 (HVO_2057), Agl12 (HVO_2059), Agl13 C60 DolPs modiﬁed by the low-salt tetrasaccharide ([M-2H] (HVO_2056), Agl14 (HVO_2058), and Agl15 (HVO_2055), ac- ion peaks detected at m/z 780.465 and 814.484, respectively)

cording to the nomenclature for archaeal N-glycosylation path- (Fig. 4A, inset). As expected, no low-salt tetrasaccharide was mbio.asm.org way components first proposed by Chaban et al. (7) and recently linked to S-layer glycoprotein Asn-498 in parent strain cells raised expanded by Meyer et al. (10) (for further discussion, see reference at the higher salinity (4A). When, however, the glycan-charged 30). Finally, the absence of Agl5-Agl15 did not compromise pools of DolP from cells deleted of aglI or aglE and grown in 3.4 M Asn-13 and Asn-83 glycosylation (Fig. S4), reflecting the specific NaCl were examined, considerably more low-salt tetrasaccharide- contributions of these proteins to the assembly of the low-salt charged DolP was detected than in the parent strain, with the tetrasaccharide. unrelated m/z 805.7 peak serving as an internal control in each The oligosaccharyltransferase AglB is not needed for Asn- case (upper panels, Fig. 4B and Fig. S6, respectively). At the same 498 glycosylation. Although Hfx. volcanii seemingly relies on two time, analysis of the S-layer glycoprotein-derived peptides ob- different pathways for the assembly of the two N-linked glycans tained from cells lacking AglI (Fig. 4B, bottom) or AglE (Fig. S6, decorating the S-layer glycoprotein, only one oligosaccharyltrans- bottom) and grown in high-salt medium revealed that low-salt ferase, AglB, has been identified in this organism (6, 31). In agree- tetrasaccharide-modified Asn-498-containing peptide was readily ment with earlier studies (13), an S-layer glycoprotein-derived detected. Asn-83-containing peptide ([M ϩ 2H]2ϩ ion peak at m/z 1,322.66) (Fig. 3A, upper) was not modified in cells deleted of aglB DISCUSSION (Fig. 3A, lower). Next, to determine whether AglB also catalyzes In Hfx. volcanii and in Hbt. salinarum, glycoproteins can be simul- transfer of the low-salt tetrasaccharide from DolP to Asn-498, taneously modified by two distinct N-linked glycans, each at- ⌬aglB cells were grown in low-salt medium and modification of tached via a different linking sugar (25, 26). Such complexity in the S-layer glycoprotein was considered by LC-ESI MS. Analysis of N-glycosylation has not been reported beyond these two archaeal the S-layer glycoprotein-derived Asn-498-containing peptide re- species. However, unlike Hbt. salinarum, where the pathway(s) of vealed peaks corresponding to the peptide modified by the com- N-glycosylation has not yet been characterized, much is known of plete low-salt tetrasaccharide ([M ϩ 2H]2ϩ ion peak at m/z the parallel process in Hfx. volcanii. In the present study, efforts 1,195.04) (Fig. 3B), as well as by the first ([M ϩ 2H]2ϩ ion peak at were directed at determining whether the same pathway em- m/z 959.96) (Fig. S5, top), the first two ([M ϩ 2H]2ϩ ion peak at ployed for generating the pentasaccharide N-linked to Hfx. volca- m/z 1,040.97) (Fig. S5, middle), and the first three ([M ϩ 2H]2ϩ nii S-layer glycoprotein Asn-13 and Asn-83 (13, 14) is also respon- ion peak at m/z 1,122.02) (Fig. S5, bottom) low-salt tetrasaccha- sible for the assembly and attachment of the low-salt ride sugars. It thus appears that AglB is not involved in the transfer tetrasaccharide N-linked to Asn-498 (25) of the same protein of the low-salt tetrasaccharide from its DolP carrier to S-layer when the cells are grown in low-salt medium. glycoprotein Asn-498. Gene deletions, combined with mass spectrometric analysis of When grown in high salt, cells lacking glycosyltransferases glycan-charged DolP and the S-layer glycoprotein, revealed that involved in Asn-13 and Asn-83 glycosylation decorate Asn-498 Agl proteins involved in the assembly of the Asn-13- and Asn-83- with the low-salt tetrasaccharide. Finally, experiments were un- linked pentasaccharide do not participate in the biosynthesis of dertaken to assess whether interplay between the two Hfx. volcanii the low-salt tetrasaccharide attached to Asn-498. Instead, the N-glycosylation pathways exists. Specifically, it was tested whether products of a distinct set of genes mediate the assembly and at- the pathway responsible for biosynthesis of the low-salt tetrasac- tachment of this glycan. Given the substantial differences in the charide could decorate Asn-498 in cells grown in medium con- compositions of the two N-linked glycans decorating the S-layer taining 3.4 M NaCl when components of the pathway responsible glycoprotein, it is not surprising that two distinct assembly path- for the assembly of the pentasaccharide N-linked to S-layer glyco- ways are involved. On the other hand, the finding that the OST protein Asn-13 and Asn-83 were missing. Accordingly, AglB is not needed for low-salt tetrasaccharide attachment to N-glycosylation of Asn-498 was considered in cells deleted of aglI S-layer glycoprotein Asn-498 is unexpected, as AglB is essential for or aglE and grown in high-salt medium. N-glycosylation of Asn-13 and Asn-83 in cells grown in either As previously reported (25), cells of the parent strain grown in high-salt or low-salt medium. In Hfx. volcanii, and indeed in Hbt.

November/December 2013 Volume 4 Issue 6 e00716-13 ® mbio.asm.org 5 Downloaded from

Kaminski et al. mbio.asm.org on November 18, 2013 - Published by mbio.asm.org

FIG 3 AglB is not involved in S-layer glycoprotein Asn-498 glycosylation. (A) ⌬aglB cells were grown in low-salt medium. LC-ESI MS analysis reveals the unmodified Asn-83-containing S-layer glycoprotein-derived peptide (upper) but not the same peptide modified by the normally bound pentasaccharide (lower). (B) LC-ESI MS analysis of the S-layer glycoprotein-derived Asn-498-containing peptide from cells lacking AglB reveals modification by the low-salt tetrasaccharide. In each panel, the position of the Asn-83 (A)- or Asn-498 (B)-containing peptide (detected or absent) is indicated, as is a schematically drawn depiction of the present or absent glycan. In the schematic drawings, the filled circles represent hexose, the filled squares represent hexuronic acid, and the open circle represents rhamnose. salinarum, a single version of AglB is the only homologue of the process that also occurs under low-salt conditions (25). More- eukaryal OST catalytic subunit, Stt3, or the bacterial OST, PglB, over, because cells lacking Agl7 contained DolP charged with a detected (6, 32). As such, a currently unidentified OST is involved nonsulfated version of the low-salt tetrasaccharide, while no Asn- in the delivery of the low-salt tetrasaccharide (and its precursors) 498-fused low-salt tetrasaccharide (or its di- or trisaccharide pre- from DolP to S-layer glycoprotein Asn-498. The same may well cursors) were detected, sulfation of the DolP-bound hexose may hold true in Hbt. salinarum, where one N-linked glycan is trans- be needed for translocation of DolP charged with a more elaborate ferred from a dolichol phosphate carrier and the second from a precursor of the low-salt tetrasaccharide or the complete glycan dolichol pyrophosphate carrier (26, 33). across the plasma membrane. Further work will be required to Based on the effects of deleting agl5-agl15 on DolP and S-layer define the precise actions of Agl5, Agl6, and Agl7, as well as the glycoprotein glycosylation, a preliminary pathway for low-salt tet- order in which they act. Agl8 and Agl9 contribute to the addition rasaccharide biogenesis can be drawn (Fig. 5). In this working of a hexose to disaccharide-charged DolP. Similarly, Agl10 to model, Agl5 and Agl6 are assigned roles in adding the linking Agl14 are involved in the appearance of the final rhamnose sugar hexose to DolP, while Agl7 contributes to the sulfation of this of the low-salt tetrasaccharide on the DolP carrier. In contrast, lipid-linked sugar. As such, the DolP-hexose seen in cells lacking cells lacking Agl15 assemble the intact low-salt tetrasaccharide on Agl5 or Agl6 may correspond to this lipid charged with the first DolP, yet no such glycan is observed on S-layer glycoprotein Asn- sugar of the pentasaccharide transferred to Asn-13 and Asn-83, a 498. This effect is consistent with Agl15 acting as a flippase, medi-

6 ® mbio.asm.org November/December 2013 Volume 4 Issue 6 e00716-13 Downloaded from

Two Hfx. volcanii N-Glycosylation Pathways mbio.asm.org on November 18, 2013 - Published by mbio.asm.org

FIG 4 Hfx. volcanii ⌬aglI cells grown in 3.4 M NaCl-containing growth medium modify DolP and S-layer glycoprotein Asn-498 with the low-salt tetrasaccharide. (A) In parent strain cells, no Asn-498-containing fragment from the S-layer glycoprotein of parent strain cells grown in high-salt medium modified by the low-salt tetrasaccharide was detected. (Inset) LC-ESI MS analysis of the low-salt tetrasaccharide-modified C55 and C60 dolichol phosphates in a total lipid extract of parent strain cells. (B) In ⌬aglI cells, low-salt tetrasaccharide attached to Asn-498 is detected (lower). (Inset) LC-ESI MS analysis of a total lipid extract from ⌬ aglI cells reveals low-salt tetrasaccharide attached to C55 and C60 dolichol phosphates. The DolP-associated peaks shown in each upper panel correspond to [M-2H]2Ϫ ions, while Asn-498 peptide-associated peaks shown in each lower panel correspond to [M ϩ 2H]2ϩ ions. In each case, the lipid- or peptide-linked low-salt tetrasaccharide is schematically depicted at the indicated expected or observed position, with the open circle corresponding to rhamnose and the filled circles corresponding to hexose.

ating the translocation of low-salt tetrasaccharide-charged DolP across the membrane. Accordingly, Agl15 shares 28% identity and 51% similarity with AglR, recently proposed to serve as (or to assist) the flippase of DolP-mannose during the assembly of the pentasaccharide added to S-layer glycoprotein Asn-13 and Asn-83 (22). Presently, one can only speculate on why the Hfx. volcanii S-layer glycoprotein is modified by the two distinct N-linked glycans under low-salt conditions but not at elevated salinity. While RNA encoding Agl5 to Agl15 was detected in cells grown at both salt levels, quantitative PCR-based studies will be necessary to determine whether comparable levels of these mRNAs are present at the different salinities. A salt concentration-related conformational change in the S-layer glycoprotein leading to exposure of Asn-498 to the low-salt tetrasaccharide N-glycosylation machinery only at the lower salinity could be imagined. The fact that this position is not modified by the Asn-13/Asn-83-linked pentasaccharide when the cells are grown at higher salinity supports this hypothesis. However, the observation that barely detectable amounts of low-salt tetrasaccharide-bound DolP are seen under FIG 5 Working model of the assembly of the low-salt tetrasaccharide deco- high-salt conditions argues that modification of Asn-498 is a ques- rating S-layer glycoprotein Asn-498 in Hfx. volcanii cells grown in 1.75 M NaCl. The sites of action of the novel Agl proteins identified in this study are tion of the availability of this glycan for transfer to this S-layer depicted. In the pathway, the vertical line corresponds to DolP, the full circles glycoprotein residue. Why cells lacking different components of correspond to hexose, and the open circle corresponds to rhamnose. The Agl the N-linked pentasaccharide biosynthetic pathway decorate Asn- proteins listed act at the cytosolic face of the plasma membrane.

November/December 2013 Volume 4 Issue 6 e00716-13 ® mbio.asm.org 7 Downloaded from

Kaminski et al.

498 with the low-salt tetrasaccharide at elevated salinity also AglD (19) using T4 DNA ligase (Promega), following excision of the awaits explanation. AglD-encoding region using NdeI and KpnI. To detect the presence of mbio.asm.org In conclusion, this study has identified components of a sec- HVO_2058 in the transformed cells, PCR amplification was performed ond Hfx. volcanii N-glycosylation pathway involved in the post- using appropriate primers (Table S2). translational modification of the S-layer glycoprotein. Indeed, ex- Mass spectrometry. LC-ESI tandem MS analysis of a total Hfx. volcanii lipid extract was performed as described by Kaminski et al. (18), while cept for the Hbt. salinarum S-layer glycoprotein, such dual LC-ESI tandem MS analysis of the S-layer glycoprotein was performed as on November 18, 2013 - Published by N-glycosylation of the same protein has not been reported else- described by Guan et al. (17). where. While it has been shown that the variant surface glycoprotein of the eukaryotic parasite Trypanosoma brucei can present two SUPPLEMENTAL MATERIAL distinct N-linked glycans, the compositions of the two oligosac- Supplemental material for this article may be found at http://mbio.asm.org charides are very similar (mannose5N-acetylglucosmine2 and /lookup/suppl/doi:10.1128/mBio.00716-13/-/DCSupplemental. Figure S1, PDF file, 0.5 MB. mannose9N-acetylglucosmine2), both relying on the same linking sugars. Moreover, both are added by isoforms of Stt3 (34, 35). Figure S2, PDF file, 1.3 MB. Thus, the finding that AglB is not the OST of the novel Hfx. vol- Figure S3, PDF file, 0.3 MB. Figure S4, PDF file, 2.4 MB. canii N-glycosylation pathway is striking. As such, an enzyme not Figure S5, PDF file, 0.3 MB. belonging to the Stt3/PglB/AglB family of OSTs is apparently ca- Figure S6, PDF file, 0.3 MB. pable of catalyzing the transfer of a lipid-linked glycan to a protein Table S1, DOCX file, 0.1 MB. target. Future efforts will need to identify this protein, as well as to Table S2, DOCX file, 0.1 MB.

address other questions, including those related to the regulation mbio.asm.org ACKNOWLEDGMENTS of this novel N-glycosylation pathway, as well as its interplay with the previously described Hfx. volcanii N-glycosylation pathway J.E. is supported by the Israel Science Foundation (grant 8/11) and the involved in adding a pentasaccharide to select Asn residues of the U.S. Army Research Office (grant W911NF-11-1-520). The mass spec- S-layer glycoprotein. trometry facility in the Department of Biochemistry of the Duke Univer- sity Medical Center and Z.G. are supported by Lipid Maps Large Scale Collaborative grant GM-069338 from the NIH. L.K. is the recipient of a MATERIALS AND METHODS Negev-Zin Associates Scholarship. Strains and growth conditions. Hfx. volcanii H53 parent strain cells (29) or the same cells deleted of aglB, aglE,oraglI (13, 15, 16, 29) were grown REFERENCES in medium containing 3.4 M NaCl (high salt) or 1.75 M NaCl (low salt), 1. Weerapana E, Imperiali B. 2006. Asparagine-linked protein glycosylation: from eukaryotic to prokaryotic systems. Glycobiology 16: 0.15 M MgSO4·7H2O, 1 mM MnCl2, 4 mM KCl, 3 mM CaCl2, 0.3% (wt/vol) yeast extract, 0.5% (wt/vol) tryptone, and 50 mM Tris-HCl, 91R–101R. pH 7.2, at 42° C (36). 2. Abu-Qarn M, Eichler J, Sharon N. 2008. Not just for Eukarya anymore: protein-glycosylation in Bacteria and Archaea. Curr. Opin. Struct. Biol. Quantitative PCR. Quantitative PCR was performed using Fast SYBR 18:544–550. green master mix (Applied Biosystems, Foster City, CA). Primers (listed 3. Dell A, Galadari A, Sastre F, Hitchen P. 2010. Similarities and differences in Table S2) were designed using Primer Express 3.0 software (PerkinEl- in the glycosylation mechanisms in prokaryotes and eukaryotes. Int. J. mer Life Sciences) and employed at a final concentration of 500 nM. For Microbiol. 2010:. PubMed148178. PCR amplification, 10-␮l reaction mixtures were subjected to 40 reaction 4. Schwarz F, Aebi M. 2011. Mechanisms and principles of N-linked protein cycles in a StepOne real-time PCR system (Applied Biosystems). Primer glycosylation. Curr. Opin. Struct. Biol. 21:576–582. efficiency was ascertained by drawing a standard curve based on 4-fold 5. Eichler J. 2013. Extreme sweetness: protein glycosylation in Archaea. Nat. serial dilutions of cDNA when primers for HVO_2058 and _2048 were Rev. Microbiol. 11:151–156. used and 10-fold serial dilutions when primers for 16S rRNA were used. 6. Kaminski L, Lurie-Weinberger MN, Allers T, Gophna U, Eichler J. 2013. Phylogenetic- and genome-derived insight into the evolutionary For expression analysis of HVO_2058 and _2048 in Hfx. volcanii H53 cells history of N-glycosylation in Archaea. Mol. Phylogenet. Evol. 67:327–339. grown in medium containing 3.4 M or 1.75 M NaCl, 40 ng of cDNA was 7. Chaban B, Voisin S, Kelly J, Logan SM, Jarrell KF. 2006. Identification used in 10-␮l PCR amplifications. For measuring the levels of the 16S of genes involved in the biosynthesis and attachment of Methanococcus rRNA housekeeping gene, 5 pg of cDNA was used in a 10-␮l reaction voltae N-linked glycans: insight into N-linked glycosylation pathways in mixture. Relative mRNA levels were quantified using the standard Archaea. Mol. Microbiol. 61:259–268. 2Ϫ⌬⌬CT formula. 8. VanDyke DJ, Wu J, Logan SM, Kelly JF, Mizuno S, Aizawa S, Jarrell KF. Gene deletion. Deletion of Hfx. volcanii genes of interest was achieved 2009. Identification of genes involved in the assembly and attachment of a using a previously described “pop-in/pop-out” approach (29, 31). To novel flagellin N-linked tetrasaccharide important for motility in the ar- amplify approximately 500-bp-long regions flanking the coding se- chaeon Methanococcus maripaludis. Mol. Microbiol. 72:633–644. 9. Maita N, Nyirenda J, Igura M, Kamishikiryo J, Kohda D. 2010. Com- quences of the various genes, appropriate primers directed against the parative structural biology of eubacterial and archaeal oligosaccharyl- upstream flanking regions and designed to introduce XhoIorKpnI and transferases. J. Biol. Chem. 285:4941–4950. HindIII sites, and against the downstream flanking regions and designed 10. Meyer BH, Zolghadr B, Peyfoon E, Pabst M, Panico M, Morris HR, to introduce BamHI and XbaI sites at the 5= and 3= of each fragment, Messner P, Schäffer C, Dell A, Albers SV. 2011. Sulfoquinovose syn- respectively, were employed (Table S2). To confirm gene deletion at the thase—an important enzyme in the N-glycosylation pathway of Sulfolobus DNA level, PCR amplification was performed using a primer directed acidocaldarius. Mol. Microbiol. 82:1150–1163. against the upstream flanking region of the gene targeted for deletion, 11. Jones GM, Wu J, Ding Y, Uchida K, Aizawa S, Robotham A, Logan SM, together with reverse primers against internal sequences of the gene in Kelly J., Jarrell KF. 2012. Identification of genes involved in the acet- question or trpA, as appropriate (Table S2). amidino group modification of the flagellin N-linked glycan of Methano- ⌬ coccus maripaludis. J. Bacteriol. 194:2693–2702. Complementation of the HVO_2058 strain was achieved upon 12. Matsumoto S, Igura M, Nyirenda J, Matsumoto M, Yuzawa S, Noda N, transformation with plasmid pWL-CBD2058, encoding HVO_2058 bear- Inagaki F, Kohda D. 2012. Crystal structure of the C-terminal globulari ing an N-terminal C. thermocellum cellulose-binding domain tag. DNA domain of oligosaccharyltransferase from Archaeoglobus fulgidus at 1.75-Å encoding HVO_2058, bearing introduced NdeI and KpnI restriction sites resolution. Biochemistry 51:4157–4166. at the 5= and 3= ends, respectively, was ligated into plasmid pWL-CBD- 13. Abu-Qarn M, Yurist-Doutsch S, Giordano A, Trauner A, Morris HR,

8 ® mbio.asm.org November/December 2013 Volume 4 Issue 6 e00716-13 Downloaded from

Two Hfx. volcanii N-Glycosylation Pathways

Hitchen P, Medalia O, Dell A, Eichler J. 2007. Haloferax volcanii AglB a Dead Sea halobacterium with a moderate salt requirement. Arch. Micro-

and AglD are involved in N-glycosylation of the S-layer glycoprotein and biol. 104:207–214. mbio.asm.org proper assembly of the surface layer. J. Mol. Biol. 374:1224–1236. 25. Guan Z, Naparstek S, Calo D, Eichler J. 2012. Protein glycosylation as an 14. Magidovich H, Yurist-Doutsch S, Konrad Z, Ventura VV, Dell A, adaptive response in Archaea: growth at different salt concentrations leads Hitchen PG, Eichler J. 2010. AglP is a S-adenosyl-L-methionine- to alterations in Haloferax volcanii S-layer glycoprotein N-glycosylation. dependent methyltransferase that participates in the N-glycosylation Environ. Microbiol. 14:743–753. pathway of Haloferax volcanii. Mol. Microbiol. 76:190–199. 26. Lechner J, Wieland F. 1989. Structure and biosynthesis of prokaryotic 15. Abu-Qarn M, Giordano A, Battaglia F, Trauner A, Hitchen PG, Morris glycoproteins. Annu. Rev. Biochem. 58:173–194. on November 18, 2013 - Published by HR, Dell A, Eichler J. 2008. Identiﬁcation of AglE, a second glycosyltrans- 27. Yurist-Doutsch S, Eichler J. 2009. Manual annotation, transcriptional ferase involved in N-glycosylation of the Haloferax volcanii S-layer glyco- analysis and protein expression studies reveal novel genes in the agl cluster protein. J. Bacteriol. 190:3140–3146. responsible for N-glycosylation in the halophilic archaeon Haloferax vol- 16. Yurist-Doutsch S, Abu-Qarn M, Battaglia F, Morris HR, Hitchen PG, canii. J. Bacteriol. 191:3068–3075. Dell A, Eichler J. 2008. aglF, aglG and aglI, novel members of a gene 28. Hartman AL, Norais C, Badger JH, Delmas S, Haldenby S, Madupu R, cluster involved in the N-glycosylation of the Haloferax volcanii S-layer Robinson J, Khouri H, Ren Q, Lowe TM, Maupin-Furlow J, Pohlsch- glycoprotein. Mol. Microbiol. 69:1234–1245. roder M, Daniels C, Pfeiffer F, Allers T, Eisen JA. 2010. The complete 17. Guan Z, Naparstek S, Kaminski L, Konrad Z, Eichler J. 2010. Distinct genome sequence of Haloferax volcanii DS2, a model archaeon. PLoS One glycan-charged phosphodolichol carriers are required for the assembly of 5:e9605. doi:10.1371/journal.pone.0009605. the pentasaccharide N-linked to the Haloferax volcanii S-layer glycopro- 29. Allers T, Ngo HP, Mevarech M, Lloyd RG. 2004. Development of tein. Mol. Microbiol. 78:1294–1303. additional selectable markers for the halophilic archaeon Haloferax volca- 18. Kaminski L, Abu-Qarn M, Guan Z, Naparstek S, Ventura VV, Raetz nii based on the leuB and trpA genes. Appl. Environ. Microbiol. 70: CR, Hitchen PG, Dell A, Eichler J. 2010. AglJ adds the ﬁrst sugar of the 943–953. N-linked pentasaccharide decorating the Haloferax volcanii S-layer glyco- 30. Eichler J, Jarrell KF, Albers S. 2013. A proposal for the naming of

protein. J. Bacteriol. 192:5572–5579. N-glycosylation pathway components in Archaea. Glycobiology 23: mbio.asm.org 19. Plavner N, Eichler J. 2008. Defining the topology of the N-glycosylation 620–621. pathway in the halophilic archaeon Haloferax volcanii. J. Bacteriol. 190: 31. Abu-Qarn M, Eichler J. 2006. Protein N-glycosylation in Archaea: defin- 8045–8052. ing Haloferax volcanii genes involved in S-layer glycoprotein glycosyla- 20. Calo D, Guan Z, Naparstek S, Eichler J. 2011. Different routes to the tion. Mol. Microbiol. 61:511–525. same ending: comparing the N-glycosylation processes of Haloferax vol- 32. Magidovich H, Eichler J. 2009. Glycosyltransferases and oligosaccharyl- canii and Haloarcula marismortui, two halophilic archaea from the Dead transferases in Archaea: putative components of the N-glycosylation path- Sea. Mol. Microbiol. 81:1166–1177. way in the third domain of life. FEMS Microbiol. Lett. 300:122–130. 21. Cohen-Rosenzweig C, Yurist-Doutsch S, Eichler J. 2012. AglS, a novel 33. Wieland F, Dompert W, Bernhardt G, Sumper M. 1980. Halobacterial component of the Haloferax volcanii N-glycosylation pathway, is a glycoprotein saccharides contain covalently linked sulphate. FEBS Lett. dolichol phosphate-mannose mannosyltransferase. J. Bacteriol. 194: 120:110–114. 6909–6916. 34. Izquierdo L, Schulz BL, Rodrigues JA, Güther ML, Procter JB, Barton 22. Kaminski L, Guan Z, Abu-Qarn M, Konrad Z, Eichler J. 2012. AglR is GJ, Aebi M, Ferguson MA. 2009. Distinct donor and acceptor specificities required for addition of the final mannose residue of the N-linked glycan of Trypanosoma brucei oligosaccharyltransferases. EMBO J. 28: decorating the Haloferax volcanii S-layer glycoprotein. Biochim. Biophys. 2650–2661. Acta 1820:1664–1670. 35. Izquierdo L, Mehlert A, Ferguson MA. 2012. The lipid-linked oligosac- 23. Yurist-Doutsch S, Magidovich H, Ventura VV, Hitchen PG, Dell A, charide donor specificities of Trypanosoma brucei oligosaccharyltrans- Eichler J. 2010. N-glycosylation in Archaea: on the coordinated actions of ferases. Glycobiology 22:696–703. Haloferax volcanii AglF and AglM. Mol. Microbiol. 75:1047–1058. 36. Mevarech M, Werczberger R. 1985. Genetic transfer in Halobacterium 24. Mullakhanbhai MF, Larsen H. 1975. Halobacterium volcanii spec. nov., volcanii. J. Bacteriol. 162:461–462.

November/December 2013 Volume 4 Issue 6 e00716-13 ® mbio.asm.org 9

Table S2 – Primers used in this study

Primer Sequence1

Gene deletion

HVO_2046uF cccggtaccAAAGCCATTTGCTGAGACG HVO_2046uR gggaagcttATGTATTCGCTTATCTGGAG HVO_2046dF cccggatccGGTAATCAACGCGGCCAAC HVO_2046dR gggtctagaCTGTTCAACAGACTCTTGTC HVO_2047uF cccctcgagCAGTAGTTGTTTGTTTGATACC HVO_2047uR gggaagcttATCATCTATTCAATTAATGTTCACC HVO_2047dF cccggatccTATGTCTAGTGACTACTCGAAC HVO_2047dR gggtctagaGATCGAGTCTTGAACTAATGTG HVO_2048uF gggctcgagGCTGCAATTGAAGCAAATGACG HVO_2048uR cccaagcttACTATACTTTGTATCCTAAGTC HVO_2048dF gggggatccGGTTCTGCTCACGTTGCTATATT HVO_2048dR ccctctagaGATTTAGAATTCTTGGTGAGATG HVO_2049uF cccctcgagTCGTGAGAAATATGATCTTG HVO_2049uR gggaagcttCAAGATCTGGGATATTTAAAATATAG HVO_2049dF cccggatccCCTTCAGACAAGTATCAGTAC HVO_2049dR gggtctagaTACGCGAAACGCTTCGGTATC HVO_2053uF gggctcgagATAACAACCAAGTTGCATCA HVO_2053uR  cccaagcttTAGCTAATCACCTTGAGTTT HVO_2053dF gggggatccCCCATAAGTGGGATTGGACC HVO_2053dR cccgagctcCAAAATCCGGTTGTACGGCT HVO_2055uF gggctcgagGATACGACCCTTCGGGCAGACGA HVO_2055uR  cccaagcttCGGTTAGCCTCCATTAGTAATCTTAC HVO_2055dF gggggatccTCTCAGTACCTCACAAAGCCGTTTAG HVO_2055dR ccctctagaGTTCTTCACGCGTGCGTACAGTG HVO_2056uF cccctcgagCCAATTCTCGGGCAAAAAAAG HVO_2056uR  gggaagcttCTTTCCTAGGGATACAATTCG HVO_2056dF cccggatccACAGCAACCTGCCAGCAGGG HVO_2056dR gggtctagaGTCCGCAATCCCGAACTGCTG HVO_2057uF cccctcgagCTCCAAGTCAACGCCGACC HVO_2057uR  gggaagcttCGATATGAGACGATACCTG HVO_2057dF cccggatccCCATGCACATCATCACCCAC HVO_2057dR gggtctagaCACCGACCAGACGATGACGC HVO_2058uF gggctcgagTCGTCTTCCGTGAACTTCCC HVO_2058uR  cccaagcttGTACTCGTTGTGTCGATGAA HVO_2058dF gggggatccTGGCTTTGTTGTCGAGGAAG HVO_2058dR ccctctagaGGACGTATCGAACTCGCGGA HVO_2059uF cccctcgagCACGTCACGAACGGCGTGCTC HVO_2059uR  gggaagcttATGCCACCAACAGAGGCAGC HVO_2059dF cccggatccATGGCTGACCAGCGCAACCAG HVO_2059dR gggtctagaGTTGGCTGTGCTGAGTGTCC HVO_2060uF cccctcgagCCTACTCGGCGACGAAAGCC HVO_2060uR  gggaagcttCTATTCGTCGTCACCGAGG HVO_2060dF cccggatccGGTTGAAAAACAATTCTTTG HVO_2060dR gggtctagaCTTGCGTTCACGTGGGTGAC HVO_2061uF  gggctcgagGCGAGCGTGTCGACGAGCAG HVO_2061uR cccaagcttGTACTGGATTCCTCCGTGTC HVO_2061dF gggggatccCGCTCGACGTAGGGGTGGAG HVO_2061dR ccctctagaCCGACCACGAGGATTCCCGAC trpAR cccgaattcTTATGTGCGTTCCGGATGCG qPCR

HVO_2048qF CTGTGTTCGAGGACGTTTCA HVO_2048qR GATCGACGCTTTTGCTTAGG HVO_2058qF AGCACCGATTCAGGAGTACG HVO_2058qR ACCGTAGACGAACGACAAGC 16SqF CGGGTTGTGAGAGCAAGAG 16SqR GGTCGAGAAAAGCGAGGAC  Complementation

HVO_2058cF gggcatATGTACGCATTCGTCACCGGC HVO_2058cR cccggtaccCTACGAGCTGTAATCGCTGAACG

1 Genomic sequence in capital letters

  



Chapter 3.6

"Phylogenetic- and genome-derived insight into the evolution of N-

glycosylation in Archaea"

Lina Kaminski, Mor N. Lurie-Weinberger, Thorsten Allers, Uri Gophna and Jerry Eichler Mol. Phylogenet. Evol. 2013b. 68:327-339

Abstract:

N-glycosylation, the covalent attachment of oligosaccharides to target protein

Asn residues, is a posttranslational modification that occurs in all three domains of life. In Archaea, the N-linked glycans that decorate experimentally characterized glycoproteins reveal a diversity in composition and content unequaled by their bacterial or eukaryal counterparts. At the same time, relatively little is known of archaeal N-glycosylation pathways outside of a handful of model strains. To gain insight into the distribution and evolutionary history of the archaeal version of this universal protein-processing event, 168 archaeal genome sequences were scanned for the presence of aglB, encoding the known archaeal oligosaccharyltransferase, an enzyme key to N-glycosylation. Such analysis predicts the presence of AglB in 166 species, with some species seemingly containing multiple versions of the protein.

Phylogenetic analysis reveals that the events leading to aglB duplication occurred at various points during archaeal evolution. In many cases, aglB is found as part of a cluster of putative N-glycosylation genes. The presence, arrangement and nucleotide composition of genes in aglB-based clusters in five species of the halophilic archaeon

Haloferax points to lateral gene transfer as contributing to the evolution of archaeal

N- glycosylation.

Molecular Phylogenetics and Evolution 68 (2013) 327–339

Contents lists available at SciVerse ScienceDirect

Molecular Phylogenetics and Evolution

journal homepage: www.elsevier.com/locate/ympev

Phylogenetic- and genome-derived insight into the evolution of N-glycosylation in Archaea ⇑ Lina Kaminski a, Mor N. Lurie-Weinberger b, Thorsten Allers c, Uri Gophna b, Jerry Eichler a, a Department of Life Sciences, Ben Gurion University, Beersheva 84105, Israel b Department of Molecular Microbiology and Biotechnology, Tel Aviv University, Tel Aviv 69978, Israel c School of Biology, University of Nottingham, Nottingham NG7 2UH, UK article info abstract

Article history: N-glycosylation, the covalent attachment of oligosaccharides to target protein Asn residues, is a post- Received 25 January 2013 translational modiﬁcation that occurs in all three domains of life. In Archaea, the N-linked glycans that Revised 23 March 2013 decorate experimentally characterized glycoproteins reveal a diversity in composition and content Accepted 26 March 2013 unequaled by their bacterial or eukaryal counterparts. At the same time, relatively little is known of Available online 6 April 2013 archaeal N-glycosylation pathways outside of a handful of model strains. To gain insight into the distribution and evolutionary history of the archaeal version of this universal protein-processing event, 168 Keywords: archaeal genome sequences were scanned for the presence of aglB, encoding the known archaeal oli- Archaea gosaccharyltransferase, an enzyme key to N-glycosylation. Such analysis predicts the presence of AglB N-glycosylation Oligosaccharyltransferase in 166 species, with some species seemingly containing multiple versions of the protein. Phylogenetic analysis reveals that the events leading to aglB duplication occurred at various points during archaeal evolution. In many cases, aglB is found as part of a cluster of putative N-glycosylation genes. The presence, arrangement and nucleotide composition of genes in aglB-based clusters in ﬁve species of the halophilic archaeon Haloferax points to lateral gene transfer as contributing to the evolution of archaeal N- glycosylation. Ó 2013 Elsevier Inc. All rights reserved.

1. Introduction Igura et al., 2008; Kelly et al., 2009; Peyfoon et al., 2010; Ng et al., 2011; Matsumoto et al., 2012; Vinogradov et al., 2012). Apart Originally thought to be a process restricted to Eukarya, it is from the two similar Methanococcus N-linked glycans, distinct pro- now clear that Bacteria and Archaea can also modify proteins via tein-bound oligosaccharides are seen in each species. Moreover, in the addition of oligosaccharides to selected Asn residues, i.e. per- both Halobacterium salinarum and Haloferax volcanii, the S-layer form N-glycosylation (Calo et al., 2010; Nothaft and Szymanski, glycoprotein is simultaneously modified by two distinct N-linked 2010; Larkin and Imperiali, 2011; Eichler, 2013). At present, under- glycans (Wieland et al., 1983; Lechner et al., 1985; Guan et al., standing of archaeal N-glycosylation lags behind that of the paral- 2012). lel process in Eukarya and Bacteria. Nonetheless, analysis of even a Given the species-specific profile of N-linked glycans that deco- limited number of archaeal glycoproteins has made it clear that rate archaeal glycoproteins, pathways of oligosaccharide assembly archaeal N-linked glycans show a diversity of content and struc- unique to a given archaeon likely exist. Indeed, in the four Agl ture that is not seen elsewhere (Schwarz and Aebi, 2011; Eichler, (archaeal glycosylation) pathways responsible for N-glycosylation 2013). To date, N-linked glycans decorating glycoproteins or repor- studied to date, namely those of the halophile Hfx. volcanii, the ter peptides from Archaeoglobus fulgidus, Halobacterium salinarum, methanogens M. voltae and M. maripaludis, and the thermoacido- Haloferax volcanii, Methanococcus maripaludis, Methanococcus vol- phile S. acidocaldarius (Calo et al., 2010; Jarrell et al., 2010; Albers tae, Methanothermus fervidus, Pyrococcus furiosus, Sulfolobus acido- and Meyer, 2011; Eichler, 2013), few common components are caldarius and Thermoplasma acidophilum have been characterized seen. The oligosaccharyltransferase (OST) AglB, responsible for (Wieland et al., 1983; Lechner et al., 1985; Kärcher et al., 1993; delivery of the assembled glycan and its precursors from a phos- Zähringer et al., 2000; Voisin et al., 2005; Abu-Qarn et al., 2007; phorylated dolichol lipid carrier to target protein Asn residues, is, however, present in each pathway. With this is mind, Magidovich and Eichler (2009) relied on the presence of aglB, encoding the only OST currently identified in Archaea, to predict the existence of a N- ⇑ Corresponding author. Address: Department of Life Sciences, Ben Gurion University, P.O. Box 653, Beersheva 84105, Israel. Fax: +972 8647 9175. glycosylation pathway in 54 of the 56 species for which complete E-mail address: [email protected] (J. Eichler). genome sequence information was available at the time.

Today, the number of publicly available archaeal genome into the prevalence, and by extension, the importance of N- sequences, including those of several phylogenetically proximal glycosylation in Archaea, may be obtained. Of the 168 genomes species, is approaching 200. This wealth of genomic data lends considered, 166 contained AglB-encoding sequences (Table 1), itself to a detailed examination of archaeal N-glycosylation from including A. fulgidus AF0329, Hfx. volcanii HVO_1530, M. voltae an evolutionary perspective. Accordingly, the examination of MVO1749, M. maripaludis MMP1424 and P. furiosus PF0156, AglB putative N-glycosylation pathway components across genome proteins all experimentally demonstrated to possess OST activity lines reported here offers novel insight into the evolution of (Chaban et al., 2006; Abu-Qarn et al., 2007; Igura et al., 2008; the archaeal version of this universal post-translational protein VanDyke et al., 2009; Matsumoto et al., 2012). In fact, AglB is modification. predicted to exist in members of all five archaeal phyla, i.e. Crenarchaeota, Euryarchaeota, Korarchaeota, Nanoarchaeota and 2. Materials and methods Thaumarchaeota, further pointing to N-glycosylation as being a common trait in Archaea. It should, however, be noted that in 2.1. Databases the vast majority of cases, neither N-glycosylation nor transcription of the predicted AglB-encoding gene has been confirmed. The list of AglB proteins, identified as containing a multi-mem- Finally, and as previously reported (Magidovich and Eichler, brane-spanning N-terminal domain and a soluble C-terminal do- 2009), no aglB sequence was detected in either Aeropyrum pernix main that includes the WWDYG consensus motif implicated in or Methanopyrus kandleri, suggesting that these species do not OST function across evolution (Yan and Lennarz, 2002; Maita perform N-glycosylation. Alternatively, given that Aeropyrum et al., 2010; Lizak et al., 2011), was obtained by scanning the fol- pernix and Methanopyrus kandleri are characterized by an atypical lowing: GT family 66 at the Carbohydrate-Active Enzymes data- gene content (Brochier et al., 2004), it is possible that a different, base (http://www.cazy.org), the Integrated Microbial Genomes – currently unrecognized OST mediates N-glycosylation in these Genome Encyclopedia of Bacteria and Archaea Genomes (IMG/ species. GEBA) (http://img.jgi.doe.gov/cgi-bin/w/main.cgi), using the term ‘EC 2.4.1.119’ as query, and the NCBI Protein Database (http:// 3.2. Multiple versions of AglB appeared throughout evolution www.ncbi.nlm.nih.gov/protein) sites, using the terms ‘Stt3’ or ‘AglB’ as query. These searches were complemented by manual In 31 of the 113 euryarchaeal species considered, and in only 2 searches of non-annotated proteins for the presence of WWDXG, of the 55 non-euryarchaeal species addressed, two or more aglB se- a relaxed form of the WWDYG motif. quences were identified. Of the 31 euryarchaeal species, 14 were methanogens (out of a total of 49 methanogens considered). In examining those methanoarchaeal species containing two or more 2.2. Phylogenetic analysis copies of aglB, no common phenotypic trait, such as an ability to grow under a given condition, is apparent. On the other hand, in The sequences of Haloferax AglB proteins were retrieved from addressing thermo- and hyperthermophilic euryarchaea, two or the IMG/GEBA website utilizing the ‘‘Gene Neighborhood’’ func- more aglB sequences were identified in all nine Thermococcus spe- tion. Homologs were aligned using MUSCLE (Edgar, 2004). The cies, in six of the seven Pyrococcus species considered, and in two of Halorubrum lacusprofundi AglB sequence served as an out-group. the three Archaeoglobus species examined. Yet, the possibility that The alignment was manually edited and ambiguously aligned posi- multiplicity of AglB in a given species is related to an elevated opti- tions were removed. The tree was then constructed utilizing the mal growth temperature is unlikely, since of the 45 crenarcheal PhyML server (http://www.atgc-montpellier.fr/phyml/)(Guindon species, all of which are thermo- or hyperthermophiles, only two, et al., 2010), using the JTT model + 4 gamma categories to approx- belonging to different genera, contain a pair of predicted AglB- imate the different substitution rates among sites, an estimation of encoding genes. invariant sites, and 100 bootstrap trials. A neighbor-joining phylo- To gain insight into the evolutionary relationship of the multi- genetic tree was generated from the list of euryarchaeal species ple versions of AglB found in euryarchaeal species, phylogenetic containing more than one copy of AglB utilizing MEGA 5 software analysis was performed (Fig. 1). The phylogenetic tree obtained as- (Tamura et al., 2011). Homologs were aligned using ClustalW (Lar- signed the multiple AglB proteins into two major groups, termed kin et al., 2007). Robustness of the tree was assessed by a bootstrap group A, including AglB sequences from the methanogens and A. test based on 500 pseudo-replicates. Bootstrap values are shown fulgidus, and group B, including Pyrococcus and Thermococcus AglB on the nodes of the tree where greater than 50%. sequences. Group A was in turn divided into two major clades (clades a and b) and an additional clade (clade c) containing the 2.3. Calculation of the codon adaptation index (CAI) and effective two Methanobacterium sp. AL-21 AglB sequences, while group B number of codons (ENC) was divided into two major clades (clades d and e). A more com- prehensive taxonomic representation of AglB phylogeny can be Calculation of CAI (Sharp and Li, 1987) and ENC (Wright, 1990) found in Fig. S1. for all clustered agl genes in five Haloferax species was performed Closer analysis of the phylogenetic tree revealed a situation utilizing Inca 2.0 (Supek and Vlahovicek, 2004). The CAI calcula- whereby the multiple versions of AglB encoded by certain species tions required manual indication of highly expressed ribosomal appeared at different times during evolution, rather than being the protein-encoding genes, which were located relying on genomic result of a single duplication event. For example, the two Methan- annotations. ocella arvoryzae AglB sequences found adjacent to each other in the genome likely appeared due to a recent gene duplication event. A 3. Results and discussion third AglB sequence from this species is assigned to cluster b, reflecting an earlier appearance of two versions of the protein 3.1. In Archaea, the gene encoding AglB is almost universally detected (Fig. 1, arrows). This claim is supported by the fact that a similar pattern of AglB distribution is seen in other species from the same Given the central role played by AglB in archaeal N-glycosyla- genus, i.e. Methanocella conradii and Methanocella paludicola.On tion, 168 publicly available archaeal genomes were scanned for the other hand, the pair of Methanocelleus marisnigri AglB proteins the presence of the encoding gene. In this manner, better insight encoded by genes found at distant positions from each other in the L. Kaminski et al. / Molecular Phylogenetics and Evolution 68 (2013) 327–339 329

Table 1 Archaeal AglB homologs.

Genus (Phyluma/Classb/Orderc/Familyd) Species AglB Acidilobus (C/Tp/Ac/Ac) A. saccharovorans ASAC_0278,0710 Caldisphaera (C/Tp/Ca/Ca) C. lagunensis Calag_0026,1336 Desulfurococcus (C/Tp/De/De) D. amylolyticus SphmelDRAFT_0373 D. fermentans Desfe_0211 D. kamchatkensis DKAM_0136 D. mucosus Desmu_0294 Ignicoccus (C/Tp/De/De) I. hospitalis Igni_0016 Ignisphaera (C/Tp/De/De) I. aggregans Igag_0094 Staphylothermus (C/Tp/De/De) S. hellenicus Shell_0596 S. marinus Smar_0223 Thermogladius (C/Tp/De/De) T. cellulolyticus TCELL_1363 Thermosphaera (C/Tp/De/De) T. aggregans Tagg_0313 Hyperthermus (C/Tp/De/Py) H. butylicus Hbut_1205 Pyrolobus (C/Tp/De/Py) P. fumarii Pyrfu_1528 Fervidicoccus (C/Tp/Fe/Fe) F. fontis FFONT_0123 Acidianus (C/Tp/Su/Su) A. hospitalis Ahos_1254 Metallosphaera (C/Tp/Su/Su) M. cuprina Mcup_0430 M. sedula Msed_1805 M. yellowstonensis MetMK1DRAFT_00024050 Sulfolobus (C/Tp/Su/Su) S. acidocaldarius Saci_1274 S. islandicus HVE10/4 SiH_1127 S. islandicus L.D.8.5 LD85_1283 S. islandicus L.S.2.15 LS215_1264 S. islandicus M.14.25 M1425_1167 S. islandicus M.16.27 M1627_1231 S. islandicus M.16.4 M164_1156 S. islandicus REY15A SiRe_1041 S. islandicus Y.G.57.14 YG5714_1163 S. islandicus Y.N.15.51 YN1551_1688 S. solfataricus 98/2 Ssol_2025 S. solfataricus P2 SSO1052 S. tokodaii ST0940 Thermofilum (C/Tp/Th/Thf) T. pendens Tpen_0640 Caldivirga (C/Tp/Th/Thp) C. maquilingensis Cmaq_0438 Pyrobaculum (C/Tp/Th/Thp) P. aerophilum PAE3030 P. arsenaticum Pars_1781 P. calidifontis Pcal_0997 P. islandicum Pisl_0431 P. oguniense Pogu_0350 Pyrobaculum sp. 1860 P186_1486 Thermoproteus (C/Tp/Th/Thp) T. neutrophilus Tneu_1689 T. tenax TTX_0519 T. uzoniensis TUZN_0151 Vulcanisaeta (C/Tp/Th/Thp) V. distributa Vdis_2064 V. moutnovskia VMUT_0472 Archaeoglobus (E/Ar/Ar/Ar) A. fulgidus AF0040,0329e,0380 A. profundus Arcpr_0726,1194 A. veneficus Arcve_0568 Ferroglobus (E/Ar/Ar/Ar) F. placidus Ferp_2437 Haladaptatus (E/H/H/H) Hap. paucihalophilus ZOD2009_20113 Halalkalicoccus (E/H/H/H) Hac. jeotgali HacjB3_10630 Haloarcula (E/H/H/H) Har. californiae HAH_00005860 Har. hispanica HAH_1202 Har. marismortui rrnAC0431 Har. sinaiiensis HAI_00022250 Har. vallismortis HAJ_00008880 Haloarcula sp. AS7094 pSCM201p1 Halobacterium (E/H/H/H) Hbt. salinarum R1 OE2548F Halobacterium sp. NRC-1 VNG1068G Halobacterium sp. DL1 HalDL1_1649 Halobiforma (E/H/H/H) Hbf. lacisalsi HlacAJ_010100009178 Haloferax (E/H/H/H) Hfx. denitrificans HAK_00032060 Hfx. mediterranei HFX_1592 Hfx. mucosum HAM_16650 Hfx. sulfurifontas HAN_00007740 Hfx. volcanii HVO_1530 Halogeometricum (E/H/H/H) Hgm. borinquense Hbor_17000 Halomicrobium (E/H/H/H) Hmc. mukohataei Hmuk_2752 Halopiger (E/H/H/H) Hpg. xanaduensis Halxa_2340 Haloquadratum (E/H/H/H) Hqr. walsbyi C23 Hqrw_3013 Hqr. walsbyi DSM 16790 HQ2681A Halorhabdus (E/H/H/H) Hrd. tiamatea HLRTI_06344 Hrd. utahensis Huta_2808 Halorubrum (E/H/H/H) Hrr. lacusprofundi Hlac_1062

(continued on next page) 330 L. Kaminski et al. / Molecular Phylogenetics and Evolution 68 (2013) 327–339

Table 1 (continued)

Genus (Phyluma/Classb/Orderc/Familyd) Species AglB Haloterrigena (E/H/H/H) Htg. turkmenica Htur_2957 Natrialba (E/H/H/H) Nab. magadii Nmag_0927 Natrinema (E/H/H/H) Nnm. pellirubrum Natpe_0008 Natronobacterium (E/H/H/H) Nbt. gregoryi Natgr_1685 Natronomonas (E/H/H/H) Nmn. pharaonis NP3720A Methanobacterium (E/Mtb/Mtb/Mtb) Methanobacterium sp. AL-21 Metbo_0534,0719 Methanobacterium sp. SWAN-1 MSWAN_1855 Methanobrevibacter (E/Mtb/Mtb/Mtb) M. ruminantium mru_0391 M. smithii ATCC 35061 Msm_0716 M. smithii DSM 2374 METSMIF1_02364 M. smithii DSM 2375 METSMIALI_01371 Methanosphaera (E/Mtb/Mtb/Mtb) M. stadtmanae Msp_0368 Methanothermobacter (E/Mtb/Mtb/Mtb) M. marburgensis MTBMA_c02090 M. thermautotrophicus MTH1623 Methanothermus (E/Mtb/Mtb/Mtm) M. fervidus Mfer_0177 Methanocaldococcus (E/Mtc/Mtc/Mtc) M. fervens Mefer_0590 M. infernus Metin_1222 M. jannaschii MJ1525 Methanocaldococcus sp. FS406-22 MFS40622_0538 M. vulcanius Metvu_0243 Methanotorris (E/Mtc/Mtc/Mtc) M. formicicus MetfoDRAFT_0029 M. igneus Metig_1797 Methanococcus (E/Mtc/Mtc/Mcc) M. aeolicus Maeo_1409 M. maripaludis C5 MmarC5_0154 M. maripaludis C6 MmarC6_1249 M. maripaludis C7 MmarC7_0669 M. maripaludis S2 MMP1424e M. maripaludis X1 GYY_07955 M. vannielii Mevan_0735 M. voltae A3 MVO_1038 M. voltae PS MVO1749e Methanothermococcus (E/Mtc/Mtc/Mcc) M. okinawensis Metok_0791 Methanocella (E, Mtm, Mtl, Mtl) M. arvoryzae LRC539,541,558 M. conradii Mtc_0182,0183,0205 M. paludicola SANAE MCP_2705,2723 Methanocorpusculum (E, Mtm, Mmb, Mcp) M. labreanum Mlab_0662 Methanoculleus (E, Mtm, Mmb, Mmc) M. marisnigri Memar_0175,2235 Methanofollis (E, Mtm, Mmb, Mmc) M. liminatans Metli_2406 Methanoplanus (E, Mtm, Mmb, Mmc) M. limicola Metlim_1216 M. petrolearius Mpet_0084,2443 Methanolinea (E, Mtm, Mmb, Mrg) M. tarda MettaDRAFT_0779 Methanoregula (E, Mtm, Mmb, Mrg) M. boonei Mboo_0249,1209 Methanosphaerula (E, Mtm, Mmb, Mrg) M. palustris Mpal_0785 Methanospirillum (E, Mtm, Mmb, Msp) M. hungatei Mhun_2859,3066,3149 Methanosaeta (E, Mtm, Msc, MSa) M. concilii MCON_1133,1444 M. harundinacea Mhar_0540,1091,1439,1730 M. thermophila Mthe_1164,1498 Methanococcoides (E, Mtm, Msc, MSr) M. burtonii Mbur_1579 Methanohalobium (E, Mtm, Msc, MSr) M. evestigatum Metev_1257 Methanohalophilus (E, Mtm, Msc, MSr) M. mahii Mmah_0123 Methanosalsum (E, Mtm, Msc, MSr) M. zhilinae Mzhil_1653 Methanosarcina (E, Mtm, Msc, MSr) M. acetivorans MA_1172,3752,3753,3754 M. barkeri Mbar_A0242,A0243,A0368 M. mazei MM_0646,0647,2210 Candidatus Haloredivivus (E/Nnh//) Candidatus Haloredivivus sp. G17 HRED_02810 Candidatus Nanosalina (E/Nnh//) Candidatus Nanosalina sp. J07AB43 J07AB43_03340 Candidatus Nanosalinarum (E/Nnh//) Candidatus Nanosalinarum J07AB56 J07AB56_11160 Pyrococcus (E/Tc/Tc/Tc) P. abyssi PAB0974,1586,2202 P. furiosus COM1 PFC_07420 P. furiosus DSM 3638 PF0156e,0411 P. horikoshii PH0242,1271 P. yayanosii PYCH_17920,19200 Pyrococcus sp. NA2 PNA2_0761,1113 Pyrococcus sp. ST04 Py04_0309,0456 Thermococcus (E/Tc/Tc/Tc) T. barophilus TERMP_00665,02078,02121 T. gammatolerans TGAM_0406,0937 T. kodakarensis TK0810,1718 T. litoralis OCC_09883, 01289, 05039 T. onnurineus TON_0775,1820 T. sibiricus TSIB_0007,0418 Thermococcus sp. 4557 GQS_05995,06090,01010 Thermococcus sp. AM4 TAM4_672,1026 Thermococcus sp. CL1 CL1_0839,0859,1904 Picrophilus (E/Tl/Tl/Pi) P. torridus PTO0786 Thermoplasma (E/Tl/Tl/Ts) T. acidophilum Ta1136 T. volcanium TVN1212 L. Kaminski et al. / Molecular Phylogenetics and Evolution 68 (2013) 327–339 331

Table 1 (continued)

Genus (Phyluma/Classb/Orderc/Familyd) Species AglB Aciduliprofundum (E///) A. boonei Aboo_0310 Candidatus Micrarchaeum (E///) Candidatus M. acidiphilum UNLARM2_0813 Candidatus Parvarchaeum (E///) Candidatus P. acidiphilum BJBARM4_0616 Candidatus P. acidophilus BJBARM5_0254 uncultured marine group II euryarchaeote MG2_1283 Candidatus Korarchaeum (K///) Candidatus K. cryptoﬁlum Kcr_1056 Nanoarchaeum (N///) N. equitans NEQ155 Cenarchaeum (T//Ce/Ce) C. symbiosum CENSYa_1939 Candidatus Nitrosoarchaeum (T//Ni/Ni) Candidatus N. koreensis MY1_0015 Candidatus N. limnia BG20 CNitlB_010100007878 Candidatus N. limnia SFB1 Nlim_2107 Nitrosopumilus (T//Ni/Ni) Candidatus N. salaria BD31_I1640 N. maritimus Nmar_0075 Nitrosopumilus sp. MY1 MY1_0015 Candidatus Caldiarchaeum (T///) Candidatus C. subterraneum CSUB_C0660 unclassiﬁed Archaea halophilic archaeon DL31 Halar_1620

Sequences obtained from the Carbohydrate-Active enZYmes Database (http://www.cazy.org/Home.html) (August, 2012), the Integrated Microbial Genomes – Genome Encyclopedia of Bacteria and Archaea Genomes (IMG/GEBA) (http://img.jgi.doe.gov/cgi-bin/geba/main.cgi) (August, 2012), UCSC Archaeal Genome Browser (http://archaea. ucsc.edu/) (August, 2012) and the NCBI Protein Database (http://www.ncbi.nlm.nih.gov/protein) (August, 2012) sites. Listed as AglB at CAZy glycosyltransferase group 66 (oligosaccharyltransfases) but lacking the WWDXG motif involved in oligosaccharyltransferase activity: HAH_0492, MSWAN_1515, MSWAN_1516, MTBMA_ c4670, MTBMA_c04680, MTH420, MTH1898, MTH1906, Mfer_0275, Mfer_0623, Mhun_2859, Mhun_3066, Mhun_3149, Mthe_1548. a Phylum: Crenarchaeota, C; Euryarchaeota, E; Korarchaeota, K; Nanoarchaeota, N; Thaumarchaeota, T. b Class: Thermoprotei, Tp; Archaeoglobi, Ar; Halobacteria, H; Methanobacteria, Mtb; Methanococci, Mtc; Methanomicrobia, Mtm; Methanopyri, Mtp; Nanohaloarchaea, Nnh; Thermococci, Tc; Thermoplasmata, Tl. c Order: Acidilobales, Ac; Caldisphaeraceae, Ca; Desulfurococcales, De; Fervidicoccales, Fe; Sulfolobales, Su; Thermoproteales, Th; Archaeoglobales, Ar; Halobacteriales, H; Methanobacteriales, Mtb; Methanococcales; Mtc; Methanocellales, Mtl; Methanomicrobiales, Mmb; Methanosarcinales, Msc; Thermococcales, Tc; Thermoplasmatales, Tl; Cenarchaeales, Ce; Nitrosopumilales, Ni; Nitrososphaerales, Nt. d Family: Acidilobaceae, Ac; Caldisphaera, Ca; Desulfurococcaceae, De; Pyrodictiaceae, Py; Fervidicoccaceae, Fe; Sulfolobaceae, Su; Thermoﬁlaceae, Thf; Thermoproteaceae, Thp; Archaeoglobaceae, Ar; Halobacteriaceae, H; Methanobacteriaceae, Mtb; Methanothermaceae, Mtm; Methanocaldococcaceae, Mtc; Methanococcaceae, Mcc; Methano- cellaceae, Mtl; Methanocorpusculaceae, Mcp; Methanomicrobiaceae, Mmc; Methanoregulaceae, Mrg; Methanospirillaceae, Msp; Methanosaetaceae, Msa; Methanosarcin- aceae, Msr; Thermococcaceae, Tc; Picrophilaceae, Pi; Thermoplasmataceae, Ts; Cenarchaeaceae, Ce; Nitrosopumilaceae, Ni; Nitrososphaeraceae, Nt. e Experimentally veriﬁed to be an oligosaccharyltransferase.

genome appeared at yet another point during evolution and clus- to clade d. Similarly, the two Methanocella arvoryzae AglB proteins tered with homologs from Methanoplanus petroleanus (Fig. 1, containing the WWDYG motif were assigned to clade a, while the arrowheads). Species-specific evolutionary patterns were also seen third AglB, in which this motif was modified to WWDDG, was as- for AglB sequences from non-methanogens. For instance, the three signed to clade b. On the other hand, both AglB proteins from Meth- Thermococcus litoralis AglB sequences also appeared at distinct anoplanus petrolearius and from Methanobacterium sp. AL-21 were points during evolution (Fig. 1, diamonds). Indeed, two of these se- assigned to the same clade, despite presenting differences in this quences cluster with their homologs from Thermococcus sibiticus motif. Likewise, the Methanosaeta harundinacea AglB sequence and Thermococcus barophilus, while the third sequence clusters where a modified WWDRG motif is found (Mhar_1439) is assigned with AglB from Thermococcus kodakarenesis. to clade b, along with two of the three additional AglB proteins predicted in this organism, each of which contains a WWDYG motif at 3.3. AglB multiplicity in a given species may carry physiological this position. significance The existence of OSTs possessing unique specificities in a single organism offers a strategy for the addition of different N-linked The presence of the multiple AglB sequences in a single species glycans in a single species. This could be tested in future in genet- could be a reflection of differences in OST substrate or target pref- ically tractable species containing multiple AglB proteins, such as erence, prevalence or availability, possibly as a function of local Methanosarcina or Thermococcales species (Leigh et al., 2011), once growth conditions. Accordingly, closer examination of the multiple proof of N-glycosylation in these species has been provided. How- versions of AglB in a given species reveals differences in the con- ever, while such differential N-glycosylation has been observed in sensus WWDYG motif. It is conceivable that these modifications Hbt. salinarum and Hfx. volcanii, where S-layer glycoproteins reflect different activities of the various versions of the protein. are simultaneously modified by two distinct N-linked glycans Accordingly, examination of the phylogenetic distribution of the (Wieland et al., 1983; Lechner et al., 1985; Guan et al., 2012), each distinct versions of AglB found in a single species often suggests species only encode a single AglB protein. Nonetheless, it remains that these proteins appeared early in evolution. AglB proteins from possible that these species contain a second OST that can no longer members of the Family Thermococcaceae (Group B) that can be dis- be recognized as an AglB ortholog, or alternatively, that relies on a tinguished on the basis of variability at the fourth position of the distinct catalytic mechanism. The fact that the two N-linked gly- consensus WWDYG catalytic motif offer such examples. Within cans in Hbt. salinarum contain different linking sugars implies the the Thermococcaceae, those AglB proteins in which Tyr is replaced existence of two OSTs employing different mechanisms of with either His or Gln were all assigned to clade e (i.e., P. furiosus catalysis. PF0411, Pyrococcus horikoshii PH1271, Thermococcus sp. 4557 GQS_05995 and GQS_06090, Thermococcus gammatolerans 3.4. Identification of putative aglB-based N-glycosylation loci TGAM_0406, Thermococcus sp. AM4 TAM4_1026, Thermococcus sp. CL-1 CL1_0839 and CL1_0859 and Thermococcus onnurineus In Hfx. volcanii, one of the few archaeal species for which de- TON_1820). By contrast, AglB proteins from the same species con- tailed information on N-glycosylation is available, all but one of taining Tyr at position four of this catalytic motif were all assigned the genes known to participate in the assembly and attachment 332 L. Kaminski et al. / Molecular Phylogenetics and Evolution 68 (2013) 327–339

Fig. 1. Phylogenetic tree of euryarchaeal AglB sequences. An alignment of 77 AglB sequences from 31 euryarchaeal species containing more than one copy of AglB was used to construct a Neighbor-Joining tree. Robustness of the tree was assessed by a bootstrap test based on 500 pseudo-replicates. Bootstrap values are shown on the nodes of the tree where greater than 50%. Each entry lists the species followed by the genome-derived name of AglB, as indicated in Table 1. The limits of the different groups and clades are marked. The arrows indicate Methanocella arvoryzae AglB sequences, while the meanings of the arrowhead and diamond symbols are provided in the text. Those AglB sequences in which the Tyr of the consensus WWDYG motif is modiﬁed are indicated by the full circles.

of a pentasaccharide to selected Asn residues of N-glycosylated 2010), mannose, is found outside this cluster. On the other hand, proteins are sequestered within an aglB-containing gene cluster no N-glycosylation gene clusters (defined as containing aglB and beginning at HVO_1517, encoding AglJ, and extending to at least three other putative N-glycosylation pathway compo- HVO_1531, encoding AglM (Yurist-Doutsch and Eichler, 2009; Yur- nent-encoding genes) are seen in M. voltae, M. maripaludis or S. aci- ist-Doutsch et al., 2010). This gene cluster also includes aglP, aglQ, docaldarius, other species where genes involved in N-glycosylation aglE, aglR, aglS, aglF, aglI and aglG. Only aglD, encoding the GT have also been identified (Chaban et al., 2006; Magidovich responsible for charging the dolichol phosphate carrier with the fi- and Eichler, 2009; VanDyke et al., 2009; Meyer et al., 2011). Hence, nal sugar of the pentasaccharide (Abu-Qarn et al., 2007; Guan et al., to assess the prevalence of aglB-based N-glycosylation gene L. Kaminski et al. / Molecular Phylogenetics and Evolution 68 (2013) 327–339 333

Table 2 Clustering of putative N-glycosylation genes around archaeal aglB homologs.

Genus (Phylum/Class/Order/Familya) Species Members of putative N-glycosylation cluster Acidilobus (C/Tp/Ac/Ac) A. saccharovorans n.d. Desulfurococcus (C/Tp/De/De) D. fermentans n.d. D. kamchatkensis n.d. D. mucosus n.d. Ignicoccus (C/Tp/De/De) I. hospitalis n.d. Ignisphaera (C/Tp/De/De) I. aggregans n.d. Staphylothermus (C/Tp/De/De) S. hellenicus n.d. S. marinus n.d. Thermogladius (C/Tp/De/De) T. cellulolyticus n.d. Thermosphaera (C/Tp/De/De) T. aggregans n.d. Hyperthermus (C/Tp/De/Py) H. butylicus n.d. Pyrolobus (C/Tp/De/Py) P. fumarii n.d. Fervidicoccus (C/Tp/Fe/Fe) F. fontis n.d. Acidianus (C/Tp/Su/Su) A. hospitalis n.d. Metallosphaera (C/Tp/Su/Su) M. cuprina Mcup_0425,0426,0427,0430 M. sedula Msed_1805,1808,1809,1810,1811,1814,1816 M. yellowstonensis MetMK1DRAFT_00024050,00024120,00024130,00024150,00024220 Sulfolobus (C/Tp/Su/Su) S. acidocaldarius n.d. S. islandicus HVE10/4 n.d. S. islandicus L.D.8.5 n.d. S. islandicus L.S.2.15 n.d. S. islandicus M.14.25 n.d. S. islandicus M.16.27 n.d. S. islandicus M.16.4 n.d. S. islandicus REY15A n.d. S. islandicus Y.G.57.14 n.d. S. islandicus Y.N.15.51 n.d. S. solfataricus 98/2 n.d. S. solfataricus P2 n.d. S. tokodaii n.d. Thermofilum (C/Tp/Th/Thf) T. pendens n.d. Caldivirga (C/Tp/Th/Thp) C. maquilingensis n.d. Pyrobaculum (C/Tp/Th/Thp) P. aerophilum n.d. P. arsenaticum n.d. P. calidifontis n.d. P. islandicum n.d. P. oguniense n.d. Pyrobaculum sp. 1860 n.d. Thermoproteus (C/Tp/Th/Thp) T. neutrophilus n.d. T. tenax n.d. T. uzoniensis n.d. Vulcanisaeta (C/Tp/Th/Thp) V. distributa n.d. V. moutnovskia n.d. Archaeoglobus (E/Ar/Ar/Ar) A. fulgidus AF0035,0038,0039,0040,0043,0044,0045/0321,0322,0323a, 0323b,0324,032,0326,0327,0328,0329 A. profundus Arcpr_1194,1195,1196,1201,1202,1203,1204,1207,1214 A. veneficus Arcve_0544,0545,0546,0552,0556,055,0562,0566,0567,0568 Ferroglobus (E/Ar/Ar/Ar) F. placidus n.d. Haladaptatus (E/H/H/H) Hap. paucihalophilus ZOD2009_20058,20063,20073,20083,20098,20113 Halalkalicoccus (E/H/H/H) Hac. jeotgali HacjB3_10595,10600,10620,10625,10630 Haloarcula (E/H/H/H) Har. californiae HAH_00005730,00005780,00005820,00005840,00005850,00005860 Har. hispanica HAH_1202,1203,1206,1208,1210,1214 Har. marismortui rrnAC0419,0421,0427,0429,0430,0431 Har. sinaiiensis HAI_00022150,00022180,0002210,0002230,0002240,00022250 Har. vallismortis HAJ_00008880,00008890,00008900,00008920 Halobacterium (E/H/H/H) Hbt. salinarum R1 OE2524R,2528R,2529F,2530F,2535R,2537F,2546F,254,2548F Halobacterium sp. NRC-1 VNG1048G,1053G,1054G,1055G,1059C,1062G,1066C,1067G,1068G Halobacterium sp. DL1 HalDL1DRAFT_1630,1631,1632,1633,1634,1639,1640,1641,1642,1643,1644,1645, 1646,1647,1649 Halobiforma (E/H/H/H) Hbf. lacisalsi HlacAJ_010100009153,010100009163,010100009168,010100009173,010100009178 Haloferax (E/H/H/H) Hfx. denitrificans HAK_000032050,00032060,000032070,000032080,000032090, 000032110,000032120,000032130,000032150 Hfx. mediterranei HFX_1580,1581,1582,1587,1591,1592 Hfx. mucosum HAM_16650,16660,16700,16750,16760,16770 Hfx. sulfurifontas HAN_00007730,00007740,00007750,00007760,00007780,00007790,00007800, 00007810,00007850,00007860,00007870 Hfx. volcanii HVO_1517b,1522b,1523b,1523.1b,1524,1525b,1526b,1527b,1528b,1529b,1530,1531b Halogeometricum (E/H/H/H) Hgm. borinquense Hbor_16990,17000,17010,17020,17030,17040,17050,17060,17070,17100,17110, 17120,17130,17140,17180,17190,17200,17210 Halomicrobium (E/H/H/H) Hmc. mukohataei Hmuk_2752,2753,2754,2756,2757,2758, Halopiger (E/H/H/H) Hpg. xanaduensis Halxa_2340,2341,2342,2344,2348,2349,2351,2352,2355,2357,2538,2361,2368, 2369,2371,2372,2379,2380,2381 Haloquadratum (E/H/H/H) Hqr. walsbyi C23 Hqrw_3012,3013,3016,3017,3021,3023,3029,3036,3040,3043,3044,3045 Hqr. walsbyi DSM 16790 HQ2680A,2681A,2682A,2683A,2686A,2687A,2691A,2692A,2694A Halorhabdus (E/H/H/H) Hrd. tiamatea n.d.

(continued on next page) 334 L. Kaminski et al. / Molecular Phylogenetics and Evolution 68 (2013) 327–339

Table 2 (continued)

Genus (Phylum/Class/Order/Familya) Species Members of putative N-glycosylation cluster Hrd. utahensis n.d. Halorubrum (E/H/H/H) Hrr. lacusprofundi Hlac_1062,1063,1065,1067,1069,1071,1073,1074,1075 Haloterrigena (E/H/H/H) Htg. turkmenica Htur_2947,2949,2954,2955,2956,2957 Natrialba (E/H/H/H) Nab. magadii Nmag_0916,0917,0922,0924,0925,0926,0927 Natrinema (E/H/H/H) Nnm. pellirubrum NatpeDRAFT_0005,0006,0007,0008 Natronobacterium (E/H/H/H) Nbt. gregoryi NatgrDRAFT_1666,1669,1670,1675,1682,1683,1684,1685 Natronomonas (E/H/H/H) Nmn. pharaonis n.d. Methanobacterium (E/Mtb/Mtb/Mtb) Methanobacterium sp. AL-21 Metbo_0719,0720,0721,0722,0723,0725,0726,0727,0729,0734 Methanobacterium sp. n.d. SWAN-1 Methanobrevibacter (E/Mtb/Mtb/Mtb) M. ruminantium n.d. M. smithii ATCC 35061 n.d. M. smithii DSM 2374 n.d. M. smithii DSM 2375 n.d. Methanosphaera (E/Mtb/Mtb/Mtb) M. stadtmanae n.d. Methanothermobacter (E/Mtb/Mtb/Mtb) M. marburgensis n.d. M. thermautotrophicus n.d. Methanothermus (E/Mtb/Mtb/Mtm) M. fervidus n.d. Methanocaldococcus (E/Mtc/Mtc/Mtc) M. fervens n.d. M. infernus n.d. M. jannaschii n.d. Methanocaldococcus sp. n.d. FS406-22 M. vulcanius n.d. Methanotorris (E/Mtc/Mtc/Mtc) M. formicicus n.d. M. igneus n.d. Methanococcus (E/Mtc/Mtc/Mcc) M. aeolicus n.d. M. maripaludis C5 n.d. M. maripaludis C6 n.d. M. maripaludis C7 n.d. M. maripaludis S2 n.d. M. maripaludis X1 n.d. M. vannielii n.d. M. voltae n.d. M. voltae n.d. Methanothermococcus (E/Mtc/Mtc/Mcc) M. okinawensis n.d. Methanocella (E, Mtm, Mtl, Mtl) M. arvoryzae LRC537,539,541,542,543,544,545,547,548,549,550,551,552,553,555,558 M. conradii Mtc_0169,0171,0172,0182,0183,0186,0187,0188,0189,0190,0191,0193,0197,0198,0199, 0201,0202,0203,0205,206 M. paludicola SANAE MCP_2704,2705,2706,2707,2708,2709,2710,2711,2714,2715,2716,2717, 2718,2719,2720,2723 Methanocorpusculum (E, Mtm, Mmb, Mcp) M. labreanum Mlab_0662,0663,664,665,666 Methanoculleus (E, Mtm, Mmb, Mmc) M. marisnigri Memar_0175,0183,0184,0185,0186,0187,0188,0189,0192 Methanofollis (E, Mtm, Mmb, Mmc) M. liminatans n.d. Methanoplanus (E, Mtm, Mmb, Mmc) M. limicola n.d. M. petrolearius n.d. Methanolinea (E, Mtm, Mmb, Mrg) M. tarda MettaDRAFT_0779,0781,0782,0783,0784,0785,0786,0787 Methanoregula (E, Mtm, Mmb, Mrg) M. boonei Mboo_0249,0250,0252,0253,0254,0255 Methanosphaerula (E, Mtm, Mmb, Mrg) M. palustris n.d. Methanospirillum (E, Mtm, Mmb, Msp) M. hungatei Mhun_2852,2853,2854,2855,2856,2857,2858,2859/ 3065,3066,3067,3072,3073,3074,3075,3076,3077,3078,3079,3080,3084,3090/3138, 3145,3147,3149,3151,3154,3161 Methanosaeta (E, Mtm, Msc, MSa) M. concilii n.d. M. harundinacea Mhar_1091,1093,1094,1095,1096,1097,1098,1099,1100,1101,1102,1103,1104,1106,1110 M. thermophila n.d. Methanococcoides (E, Mtm, Msc, MSr) M. burtonii Mbur_1579,1581,1582,1583,1584,1585,1586,1587,1590,1593,1594,1597,1603,1604,1605, 1607,1608,1612,1613,1615,1617 Methanohalobium (E, Mtm, Msc, MSr) M. evestigatum Metev_1236,1237,1242,1244,1250,1252,1253,1254,1255,1257 Methanohalophilus (E, Mtm, Msc, MSr) M. mahii n.d. Methanosalsum (E, Mtm, Msc, MSr) M. zhilinae Mzhil_1638,1639,1640,1641,1642,1643,1645,1648,1649,1651,1652,1653,1655,1656 Methanosarcina (E, Mtm, Msc, MSr) M. acetivorans MA_1172,1173,1174,1175,1176,1177,1179,1180,1181,1183,1184,1185,1186,1187/ 3752,3753,3754,3755,3756,3757,3758,3764,3766,3767,3769a, 3769b,3777, 3778,3779,3780,3781 M. barkeri Mbar_A0229,A0230,A0231,A0232, A0233,A0234,A0235,A0236,A0237, A0238,A0239,A0240,A0241,A0242, A0243/A0366,A0368,A0369,A0373,A0374,A0375 M. mazei MM_0646,0647,0648,0649,0650,0651,0652,0653,0654,656,657,658,659,660/ 2208,2210,2213,2214,2215,2216,2217,2221,2222,2223 Candidatus Haloredivivus (E/Nnh//) Candidatus Haloredivivus sp. HRED_02640,02670,02680,02720,02810 G17 Candidatus Nanosalina (E/Nnh//) Candidatus Nanosalina sp. J07AB43_03180,03190,03200,03210,03240,03270,03310,03320,03330,03340 J07AB43 Candidatus Nanosalinarum (E/Nnh//) Candidatus Nanosalinarum J07AB56_11160,11200,11210,11240,11250 J07AB56 Pyrococcus (E/Tc/Tc/Tc) P. abyssi PAB1411,1410,1409,0973,0974/0783,0784,0785,0787,0789,0790.1nn, 0973,0795,0796,1587,1586 L. Kaminski et al. / Molecular Phylogenetics and Evolution 68 (2013) 327–339 335

Table 2 (continued)

Genus (Phylum/Class/Order/Familya) Species Members of putative N-glycosylation cluster P. furiosus COM1 n.d. P. furiosus DSM 3638 n.d. P. horikoshii n.d. P. yayanosii PYCH_17860,17870,17880,17900,17910,17920 Pyrococcus sp. NA2 PNA2_1113,1114,1115,1120,1121 Pyrococcus sp. ST04 Py04_0454,0455,0456,0457,0461,0465 Thermococcus (E/Tc/Tc/Tc) T. barophilus TERMP_02119,02121,02122,02123,02124,02129/ 2078,2079,2080,2084,2089,2091,2094,2096,2097,2099,2100 T. gammatolerans n.d. T. kodakarensis TK1708,1711,1712,1713,1714,1715,1716,1717,1718,1719,1720,1721, 1722,1723,1725,1731,1732,1733 T. litoralis OCC_01289,01319,01324,01329,01334,01354/05039,05049,05054,05059,05064 T. onnurineus TON_1818,1819,1820,1821,1822,1823 T. sibiricus TSIB_2044,2045,2047,2048,2049,2050,2054,2059,2061,0003,0004,0005,0006,0007 Thermococcus sp. 4557 GQS_05950,05955,05960,05975,05995,06000,0600/06075,06080,06085,06090 Thermococcus sp. AM4 TAM4_1088,1094,1026,1040 Thermococcus sp. CL1 CL1_0827,0828,0830,0834,0838,0839,0840,0841/0850,0856,0857,0859 Picrophilus (E/Tl/Tl/Pi) P. torridus n.d. Thermoplasma (E/Tl/Tl/Ts) T. acidophilum n.d. T. volcanium n.d. Aciduliprofundum (E///) A. boonei n.d. Candidatus Micrarchaeum (E///) Candidatus M. acidiphilum n.d. Candidatus Parvarchaeum (E///) Candidatus P. acidiphilum n.d. Candidatus P. acidophilus n.d. uncultured marine group II n.d. euryarchaeote Candidatus Korarchaeum (K///) Candidatus K. cryptoﬁlum n.d. Nanoarchaeum (N///) N. equitans n.d. Cenarchaeum (T//Ce/Ce) C. symbiosum A n.d. Candidatus Nitrosoarchaeum (T//Ni/Ni) Candidatus N. koreensis n.d. Candidatus N. limnia BG20 n.d. Candidatus N. limnia SFB1 n.d. Nitrosopumilus (T//Ni/Ni) Candidatus N. salaria n.d. N. maritimus n.d. Nitrosopumilus sp. MY1 n.d. Candidatus Caldiarchaeum (T///) Candidatus C. subterraneum n.d. unclassiﬁed Archaea halophilic archaeon DL31 Halar_1591,1600,1601,1610,1611,1612,1613,1615,1616,1620

Glycosylation-related annotation at the Integrated Microbial Genomes – Genome Encyclopedia of Bacteria and Archaea Genomes (IMG/GEBA) (http://img.jgi.doe.gov/cgi-bin/ geba/main.cgi) (August, 2012), UCSC Archaeal Genome Browser (http://archaea.ucsc.edu/) (August, 2012) and the NCBI Protein Database (http://www.ncbi.nlm.nih.gov/ protein) (August, 2012) sites. Cluster is deﬁned as including AglB and at least 3 other putative proteins involved in glycosylation; AglB in bold. n.d., not detected. a The abbreviations used from the different phyla, classes, orders, and families are provided in the legend to Table 1. b Sequences other than AglB experimentally conﬁrmed as participating in N-glycosylation.

clustering across the Archaea, those regions down- and upstream hyperthemophilic euryarchaeota, aglB-based glycosylation gene of genes annotated as encoding AglB were examined (Table 2). clustering was seen in the three Archaeoglobus species, in all In the 45 crenarcheal species considered, aglB-based gene clus- Thermococcus species apart from T. gammatolerans, and in four of tering was only observed in the three species belonging to the the seven Pyrococcus species. No aglB-based clustering was seen Genus Metallosphaera. Given the broad geographic distribution of in the Korarchaeota, Nanoarchaeota or Thaumarchaeota. Metallosphaera cuprina (sulfuric hot spring in Tengchong, Yunnan, In most cases where aglB-based glycosylation gene clustering China; Liu et al., 2011), Metallosphaera sedula (Thermal pond, Pisc- was observed, aglB itself corresponds to one edge of the cluster. iarelli Solfatara, Naples, Italy; Huber et al., 1989) and Metallosphae- In a limited number of cases, an additional glycosylation gene ra yellowstonensis (acidic geothermal springs in Yellowstone adjacent to aglB serves this role. Where multiple AglB sequences National Park; Kozubal et al., 2008), it would appear that N-glyco- are found, some species presented each aglB in a glycosylation sylation gene clustering occurred prior to division of an ancestor gene clusters, others organized only some of the multiple AglB se- into the three species. In the Euryarchaeota, aglB-based glycosyla- quences into such clusters, while yet other species containing tion gene clustering was detected in 26 of the 29 available haloar- multiple AglB sequences did not cluster N-glycosylation genes chaeal genomes, with only the two Halorhabdus species and around aglB at all. In each of the three Methanocella species, the Natronomonas pharaonis not presenting such an arrangement. This multiple versions of AglB were all found in a common gene is not unexpected, since gene clusters appear to be better con- cluster. served in haloarchaea than other archaeal groups (Berthon et al., Finally, the distribution of genes known or believed to mediate 2008). In the 49 methanoarchaeal species examined, aglB-based N-glycosylation in other Archaea suggests that N-glycosylation glycosylation gene clustering was observed largely along genus gene clusters not anchored by aglB may also exist. For example, lines, with some genera in a given family displaying aglB-based N-glycosylation roles have been demonstrated for the products of glycosylation gene clustering and others in the same family M. maripaludis MMP1079-MMP1088, while AglB is encoded by not. Indeed, even within a given methanoarchaeal genus, only MMP1424 (Chaban et al., 2006, 2009; Shams-Eldin et al., 2008; some species presented such gene clustering. In the thermo- and Jones et al., 2012). 336 L. Kaminski et al. / Molecular Phylogenetics and Evolution 68 (2013) 327–339

Fig. 2. Schematic representation of aglB-based gene clusters in ﬁve Haloferax species. The positions of agl genes in Hfx. volcanii and their homologs in Hfx. denitriﬁcans, Hfx. mediterranei, Hfx. mucosum and Hfx. sulfurifontas are indicated, as are those of other glycosylation-related genes. The genes are arbitrarily drawn in terms of size. In those species where the orientation of the gene cluster is opposite to that in Hfx. volcanii, a double-headed arrow is found next to the species name. The legend describes the meaning of the coloring scheme employed.

3.5. Evolutionary insight into N-glycosylation in Haloferaxspecies and other sugar-processing proteins not found in the comparable derived from gene cluster comparison and gene content Hfx. volcanii cluster. Thus, based on the composition of the Hfx. denitrificans aglB-based gene cluster, it can be predicted that N-linked To demonstrate how the phenomenon of aglB-based glycosyla- glycans in this species will be highly similar if not identical to the tion gene clustering can be used for making predictions related to N-linked pentasaccharide decorating glycoproteins in Hfx. volcanii. N-glycosylation in a given species and to gain insight into the evo- By the same reasoning, one would expect a somewhat different N- lution of this post-translational modification, the five Haloferax glycan in Hfx. sulfurifontas. In considering the identical aglB-based species for which genomic information is presently available were gene clusters seen in Hfx. mediterranei and Hfx. mucosum, not only considered. In terms of their geography, the five species are found are far fewer glycosylation-related genes observed, the few homo- distally of one another, with Hfx. denitrificans originally having logs of Hfx. volcanii Agl protein-encoding genes are distributed dif- been isolated from a saltern in California, USA (Tomlinson et al., ferently than in the Hfx. volcanii cluster. As such, the N-glycans 1986), Hfx. mediterranei from a saltern near Alicante, Spain (Rodri- predicted to decorate Hfx. mediterranei and Hfx. mucosum glycopro- guez-Valera et al., 1980), Hfx. mucosum from Shark Bay, Australia teins are expected to be identical, yet significantly differing from (Allen et al., 2008), Hfx. sulfurifontis from a sulfur spring in Okla- what is found in the other Haloferax strains. homa, USA (Elshahed et al., 2004) and Hfx. volcanii from the Dead The organization of aglB and other agl genes within each cluster Sea (Mullakhanbhai and Larsen, 1975). offers evolutionary insight into the N-glycosylation process. The At present, the pathway of N-glycosylation has been delineated similarities of the aglB-based gene clusters in Hfx. volcanii, Hfx. in Hfx. volcanii, based on a series of genetic and biochemical studies. In the Agl pathway in this species, aglJ, aglG, aglI and aglE encode GTs that sequentially add four nucleotide-activated sugars to a common dolichol phosphate carrier on the inner face of the plasma membrane (Abu-Qarn et al., 2008; Plavner and Eichler, 2008; Yurist-Doutsch et al., 2008; Guan et al., 2010; Kaminski et al., 2010). Once the tetrasaccharide-bearing dolichol phosphate has been translocated across the membrane, AglB delivers the glycan to target protein Asn residues (Abu-Qarn et al., 2007). At the same time, the only N-glycosylation pathway component encoded by a gene outside the aglB-based glycosylation gene cluster, AglD, adds the final sugar of the N-linked pentasaccharide, mannose, to a distinct dolichol phosphate (Abu-Qarn et al., 2007; Yurist-Doutsch and Eichler, 2009; Guan et al., 2010). Mannose-charged dolichol phosphate is ‘flipped’ across the membrane in a process involving AglR (Kaminski et al., 2012), at which point AglS delivers the mannose to the Asn-bound tetrasaccharide (Cohen-Rosenzweig et al., 2012). In additon, AglF, AglM and AglP serve various sugar processing roles (Magidovich et al., 2010; Yurist-Doutsch et al., 2010). Examination of the genomes of Hfx. denitrificans, Hfx. mediterranei, Hfx. mucosum and Hfx. sulfurifontas reveals aglB-based gene clusters containing homologs to many of the Hfx. volcanii agl genes (Fig. 2). For example, the Hfx. denitrificans aglB-based gene cluster is almost identical to its Hfx. volcanii counterpart, except for the presence of transposases in the latter. The Hfx. sulfurifontas aglB- Fig. 3. Phylogenetic tree of Haloferax AglB proteins. The phylogenetic relationships based gene cluster also contains several homologs to Hfx. volcanii of AglB from Hfx. volcanii, Hfx. denitrificans, Hfx. mediterranei, Hfx. mucosum and Hfx. agl sequences, albeit differently arranged. In addition, the Hfx. sul- sulfurifontas is presented. The Hrr. lacusprofundi AglB sequence served as an out- furifontas aglB-based gene cluster contains sequences encoding GTs group. Numbers represent the percent of bootstrap support for each node. L. Kaminski et al. / Molecular Phylogenetics and Evolution 68 (2013) 327–339 337

Table 3 G + C content of Haloferax agl genes.

Hfx. volcanii Hfx. denitriﬁcans Hfx. sulfurifontis Hfx. mediterranei Hfx. mucosum aglJa 0.61 0.60 0.62 0.62 0.62 aglP 0.45 0.44 –b –– aglQ 0.48 0.47 – – – aglE 0.47 0.47 0.62 – – aglR 0.47 0.46 0.49 0.61 0.62 aglS 0.41 0.41 – – – aglF 0.55 0.54 0.54 – – aglI 0.58 0.58 0.51 0.64 0.64 aglG 0.54 0.56 – – – aglB 0.63 0.62 0.62 0.62 0.63 aglM 0.66 0.65 0.67 – – aglD 0.70 0.71 0.71 0.64 0.67 Genomec 0.62 0.66 0.66 0.60 0.62

a Genes are listed in the order they appear in the Hfx. volcanii agl gene cluster, except for aglD, which is found elsewhere in the genome. b Gene not detected. c G + C content as listed at http://img.jgi.doe.gov/cgi-bin/w/main.cgi.

Table 4 Codon usage bias in Haloferax agl genes.

Hfx. volcanii Hfx. denitriﬁcans Hfx. sulfurifontis Hfx. mediterranei Hfx. mucosum aglJa 40.89b 41.82 38.74 35.79 32.89 0.506c 0.4802 0.5435 0.6255 0.7186 d aglP 51.25 51.54 – –– 0.1748 0.1551 aglQ 56.37 55.82 –– – 0.1718 0.1624 aglE 56.37 55.30 40.76 – – 0.1831 0.1628 0.5159 aglR 57.21 55.91 56.22 39.60 41.30 0.1806 0.1658 0.2015 0.5275 0.5337 aglS 56.37 49.76 –– – 0.1718 0.1246 aglF 46.98 48.19 50.83 –– 0.4565 0.4065 0.3378 aglI 47.67 46.94 56.74 34.16 33.05 0.358 0.3436 0.2089 0.6864 0.7078 aglG 43.91 42.63 –– – 0.4306 0.4613 aglB 38.28 38.84 38.01 37.72 34.97 0.5495 0.5371 0.5328 0.6469 0.6948 aglM 33.90 34.35 32.51 – – 0.6683 0.61 0.675 Genome 33.78 34.17 34.33 42.41 40.09 0.6585 0.6511 0.6494 0.5533 0.5974 s.d.e ±5.95 ±6.64 ±6.67 ±7.06 ±6.28 ±0.1199 ±0.1293 ±0.1284 ±0.1121 ±0.0990

a Genes are listed in the order they appear in the Hfx. volcanii agl gene cluster. b ENC value – values between one and two standard deviations higher than the genomic average are in bold, while values more than two standard deviations higher than the genomic average are in bold and underlined. c CAI value – values between one and two standard deviations lower than the genomic average are in bold, while values more than two standard deviations lower than the genomic average are in bold and underlined. d Gene not detected. e s.d. = standard deviation. denitrificans and Hfx. sulfurifontas and in Hfx. mediterranei and Hfx. volcanii aglD and its Hfx. denitrificans and Hfx. sulfurifontas homo- mucosum are in agreement with the grouping of these species into logs, this region was expanded to include the 9 upstream and the separate clades, based on genomic segment loss and gain studies 28 downstream genes (not shown). (Lynch et al., 2012). AglB protein phylogeny (Fig. 3) is congruent At the same time, variations in the composition of agl genes in with the Haloferax tree generated by this earlier study. At the same the aglB-based clusters of the different Haloferax strains, together time, homologs of Hfx. volcanii aglD, the only component of the N- with the concept that differences in N-linked glycosylation could glycosylation pathway in this species found outside the aglB-based provide adaptive advantages, raise the possibility that lateral gene gene cluster, were detected in Hfx. denitrificans (HAK_00016980), transfer (LGT) played a role in the evolution of AglB-based N-glyco- Hfx. mediterranei (HAL_00010870), Hfx. mucosum sylation in this genus. With the exception of Haloquadratum wal- (HAM_00012150) and Hfx. sulfurifontas (HAN_00024650). In each sbyi, haloarchaea are characterized by a high genomic G + C case, the identical six downstream and three upstream genes content (typically 65% G + C) (Hartman et al., 2010). The G + C con- bordered this GT-encoding gene. However, in the case of the Hfx. tents of aglB and aglM tend to resemble the genomic average. This 338 L. Kaminski et al. / Molecular Phylogenetics and Evolution 68 (2013) 327–339 is not the case for other agl genes, which, in several species, present facets of the archaeal version of this universal protein-processing a highly unusual base composition (Table 3). For instance, several event. of the clustered genes encoding Agl homologs in Hfx. volcanii, Hfx. denitrificans and Hfx. sulfurifontas are A + T-rich, relative to Acknowledgments the rest of the genome (i.e. a difference of P10% from the genomic mean), indicative of fairly recent horizontal acquisition. Further- The authors thank Sam Haldenby for his early contributions to more, these genes also have extremely high effective number of co- this project. JE is supported by grants from the Israel Science Foun- dons (ENC) values and low codon adaptation index (CAI) values, dation (30/07) and the US Army Research Office (W911NF-11-1- indicative of the use of codons that are rare in the respective gen- 520). UG is supported by grants from the Israel Science Foundation omes (Table 4). It is noteworthy that the CAI of aglB is substantially (201/12) and the German-Israeli Project Cooperation (DIP). TA has higher in Hfx. mediterranei and Hfx. mucosum than in the other spe- been supported by a Royal Society University Research Fellowship. cies, implying higher levels of gene expression. Combined with the LK is the recipient of a Negev-Zin Associates Scholarship. fact that the other N-glycosylation genes in these two genomes appear to be ancestral rather than recently acquired, it would appear that N-glycosylation is a more fundamental trait of Hfx. mediterra- Appendix A. Supplementary material nei and Hfx. mucosum. Additionally, there is high within-cluster variation in nucleotide composition in the agl clusters of Hfx. volca- Supplementary data associated with this article can be found, in nii and Hfx. denitrificans, indicating that agl genes were recruited the online version, at http://dx.doi.org/10.1016/j.ympev.2013.03. from different sources, in agreement with the protein-based phy- 024. logenies of these genes, which show conflicting evolutionary histo- ries (Fig. S2). References

Abu-Qarn, M., Yurist-Doutsch, S., Giordano, A., Trauner, A., Morris, H.R., Hitchen, P., Medalia, O., Dell, A., Eichler, J., 2007. Haloferax volcanii AglB and AglD are 4. Conclusions involved in N-glycosylation of the S-layer glycoprotein and proper assembly of the surface layer. J. Mol. Biol. 374, 1224–1236. Abu-Qarn, M., Giordano, A., Battaglia, F., Trauner, A., Hitchen, P.G., Morris, H.R., Dell, In 1976, the Hbt. salinarum S-layer glycoprotein became the first A., Eichler, J., 2008. Identification of AglE, a second glycosyltransferase involved non-eukaryal N-glycosylated protein reported (Mescher and in N-glycosylation of the Haloferax volcanii S-layer glycoprotein. J. Bacteriol. Strominger, 1976). Over the next fifteen years, advances in describ- 190, 3140–3146. Albers, S.V., Meyer, B.H., 2011. The archaeal cell envelope. Nat. Rev. Microbiol. 9, ing both the structures of glycans N-linked to archaeal glycopro- 414–426. teins and archaeal N-glycosylation pathways were made. More Allen, M.A., Goh, F., Leuko, S., Echigo, A., Mizuki, T., Usami, R., Kamekura, M., Neilan, recently, the availability of complete genome sequences, together B.A., Burns, B.P., 2008. Haloferax elongans sp. nov. and Haloferax mucosum sp. nov., isolated from microbial mats from Hamelin Pool, Shark Bay, Australia. Int. with the development of appropriate molecular tools and tech- J. Syst. Evol. Microbiol. 58, 798–802. niques led to renewed interest in this topic. In the last decade, con- Berthon, J., Cortez, D., Forterre, P., 2008. Genomic context analysis in Archaea siderable progress has been made in addressing genes and proteins suggests previously unrecognized links between DNA replication and translation. Genome Biol. 9, R71. involved in N-glycosylation in several species. Today, alongside Brochier, C., Forterre, P., Gribaldo, S., 2004. Archaeal phylogeny based on proteins of such efforts being conducted at the molecular level, insight into the transcription and translation machineries: tackling the Methanopyrus the archaeal version of this universal post-translational modifica- kandleri paradox. Genome Biol. 5, R17. Calo, D., Kaminski, L., Eichler, J., 2010. Protein glycosylation in Archaea: sweet and tion can now be gleaned at the genome level. extreme. Glycobiology 20, 1065–1076. As discussed here, virtually all Archaea encode for components Chaban, B., Voisin, S., Kelly, J., Logan, S.M., Jarrell, K.F., 2006. Identification of genes of a N-glycosylation pathway, pointing to such protein processing involved in the biosynthesis and attachment of Methanococcus voltae N-linked as being a common event in this life form. Moreover, even though glycans: insight into N-linked glycosylation pathways in Archaea. Mol. Microbiol. 61, 259–268. relatively few examples have been experimentally characterized, it Chaban, B., Logan, S.M., Kelly, J.F., Jarrell, K.F., 2009. AglC and AglK are involved in is already abundantly clear that archaeal N-glycosylation involves biosynthesis and attachment of diacetylated glucuronic acid to the N-glycan in more variety in terms of sugars, glycan structures, and by exten- Methanococcus voltae. J. Bacteriol. 191, 187–195. Cohen-Rosenzweig, C., Yurist-Doutsch, S., Eichler, J., 2012. AglS, a novel component sion, biosynthetic pathways than seen elsewhere. By focusing on of the Haloferax volcanii N-glycosylation pathway, is a dolichol phosphate- a single component of the archaeal N-glycosylation pathway, it mannose mannosyltransferase. J. Bacteriol. 194, 6909–6916. was shown that gene duplication and modification had occurred Edgar, R.C., 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucl. Acids Res. 32, 1792–1797. at numerous different points during evolution. Furthermore, com- Eichler, J., 2013. Extreme sweetness: protein glycosylation in Archaea. Nat. Rev. parison of the organization and content of N-glycosylation genes in Microbiol 11, 151–156. five members of the same genus revealed that substantial LGT had Elshahed, M.S., Savage, K.N., Oren, A., Gutierrez, M.C., Ventosa, A., Krumholz, L.R., 2004. Haloferax sulfurifontis sp. nov., a halophilic archaeon isolated from a occurred over the course of time. sulfide- and sulfur-rich spring. Int. J. Syst. Evol. Microbiol. 54, 2275–2279. Despite advances made in deciphering pathways of archaeal Guan, Z., Naparstek, S., Kaminski, L., Konrad, Z., Eichler, J., 2010. Distinct glycan- N-glycosylation, numerous unanswered questions remain. For in- charged phosphodolichol carriers are required for the assembly of the pentasaccharide N-linked to the Haloferax volcanii S-layer glycoprotein. Mol. stance, one can ask what species-specific changes allow the archa- Microbiol. 78, 1294–1303. eal oligosaccharyltransferase, AglB, to accommodate such a wide Guan, Z., Naparstek, S., Calo, D., Eichler, J., 2012. Protein glycosylation as an adaptive range of glycan structures. Do Archaea encountering similar envi- response in Archaea: growth at different salt concentrations leads to alterations ronmental extremes decorate their proteins with similar N-linked in Haloferax volcanii S-layer glycoprotein N-glycosylation. Environ. Microbiol. 14, 743–753. glycans? How common is the ability to modify N-glycosylation in Guindon, S., Dufayard, J.F., Lefort, V., Anisimova, M., Hordijk, W., Gascuel, O., 2010. response to changing surroundings, a phenomenon recently ob- New algorithms and methods to estimate maximum-likelihood phylogenies: served in Hfx. volcanii? As new species appeared, did N-glycosyla- assessing the performance of PhyML 3.0. Syst. Biol. 59, 307–321. Hartman, A.L., Norais, C., Badger, J.H., Delmas, S., Haldenby, S., Madupu, R., tion change at the same rate? Finally, one can ask whether it will Robinson, J., Khouri, H., Ren, Q., Lowe, T.M., Maupin-Furlow, J., Pohlschroder, become possible to describe the composition of the N-linked gly- M., Daniels, C., Pfeiffer, F., Allers, T., Eisen, J.A., 2010. The complete genome cans decorating archaeal glycoproteins based on their glycosyla- sequence of Haloferax volcanii DS2, a model archaeon. PLoS One 5, e9605. Huber, G., Spinnler, C., Gambacorta, A., Stetter, K.O., 1989. Metallosphaera sedula gen. tion gene content. Examining archaeal N-glycosylation from the nov. and sp. nov. represents a new genus of aerobic, metal-mobilizing, genomic perspective will help address these and elucidate other thermoacidophilic archaebacteria. Syst. Appl. Microbiol. 12, 38–47. L. Kaminski et al. / Molecular Phylogenetics and Evolution 68 (2013) 327–339 339

Igura, M., Maita, N., Kamishikiryo, J., Yamada, M., Obita, T., Maenaka, K., Kohda, D., Mullakhanbhai, M.F., Larsen, H., 1975. Halobacterium volcanii spec. nov., a Dead Sea 2008. Structure-guided identification of a new catalytic motif of halobacterium with a moderate salt requirement. Arch. Microbiol. 104, 207– oligosaccharyltransferase. EBMO J. 27, 234–243. 214. Jarrell, K.F., Jones, G.M., Kandiba, L., Nair, D.B., Eichler, J., 2010. S-layer glycoproteins Ng, S.Y., Wu, J., Nair, D.B., Logan, S.M., Robotham, A., Tessier, L., Kelly, J.F., Uchida, K., and flagellins: reporters of archaeal posttranslational modifications. Archaea. Aizawa, S., Jarrell, K.F., 2011. Genetic and mass spectrometry analyses of the pii: 612948. unusual type IV-like pili of the archaeon Methanococcus maripaludis. J. Bacteriol. Jones, G.M., Wu, J., Ding, Y., Uchida, K., Aizawa, S.I., Robotham, A., Logan, S.M., Kelly, 193, 804–814. J., Jarrell, K.F., 2012. Identification of genes involved in the acetamidino group Nothaft, H., Szymanski, C.M., 2010. Protein glycosylation in bacteria: sweeter than modification of the flagellin N-linked glycan of Methanococcus maripaludis.J. ever. Nat. Rev. Microbiol. 8, 765–778. Bacteriol. 194, 2693–2702. Peyfoon, E., Meyer, B., Hitchen, P.G., Panico, M., Morris, H.R., Haslam, S.M., Albers, Kaminski, L., Abu-Qarn, M., Guan, Z., Naparstek, S., Ventura, V.V., Raetz, C.R., S.V., Dell, A., 2010. The S-layer glycoprotein of the crenarchaeote Sulfolobus Hitchen, P.G., Dell, A., Eichler, J., 2010. AglJ adds the first sugar of the N-linked acidocaldarius is glycosylated at multiple sites with chitobiose-linked N- pentasaccharide decorating the Haloferax volcanii S-layer glycoprotein. J. glycans. Archaea. pii: 754101. Bacteriol. 192, 5572–5579. Plavner, N., Eichler, J., 2008. Defining the topology of the N-glycosylation Kaminski, L., Guan, Z., Abu-Qarn, M., Konrad, Z., Eichler, J., 2012. AglR is required for pathway in the halophilic archaeon Haloferax volcanii. J. Bacteriol. 190, addition of the final mannose residue of the N-linked glycan decorating the 8045–8052. Haloferax volcanii S-layer glycoprotein. Biochim. Biophys. Acta 1820, 1664– Rodriguez-Valera, F., Ruiz-Berraquero, F., Ramos-Cormenzana, A., 1980. Isolation of 1670. extremely halophilic bacteria able to grow in defined inorganic media with Kärcher, U., Schröder, H., Haslinger, E., Allmaier, G., Schreiner, R., Wieland, F., single carbon sources. J. Gen. Microbiol. 119, 535–538. Haselbeck, A., König, H., 1993. Primary structure of the heterosaccharide of the Schwarz, F., Aebi, M., 2011. Mechanisms and principles of N-linked protein surface glycoprotein of Methanothermus fervidus. J. Biol. Chem. 268, 26821– glycosylation. Curr. Opin. Struct. Biol. 21, 576–582. 26826. Shams-Eldin, H., Chaban, B., Niehus, S., Schwarz, R.T., Jarrell, K.F., 2008. Kelly, J., Logan, S.M., Jarrell, K.F., VanDyke, D.J., Vinogradov, E., 2009. A novel N- Identification of the archaeal alg7 gene homolog (encoding N- linked flagellar glycan from Methanococcus maripaludis. Carbohydr. Res. 344, acetylglucosamine-1-phosphate transferase) of the N-linked glycosylation 648–653. system by cross-domain complementation in Saccharomyces cerevisiae.J. Kozubal, M., Macur, R.E., Korf, S., Taylor, W.P., Ackerman, G.G., Nagy, A., Inskeep, Bacteriol. 190, 2217–2220. W.P., 2008. Isolation and distribution of a novel iron-oxidizing crenarchaeon Sharp, P.M., Li, W.H., 1987. The codon Adaptation Index-a measure of directional from acidic geothermal springs in Yellowstone National Park. Appl. Environ. synonymous codon usage bias, and its potential applications. Nucl. Acids Res. Microbiol. 74, 942–949. 15, 1281–1295. Larkin, A., Imperiali, B., 2011. The expanding horizons of asparagine-linked Supek, F., Vlahovicek, K., 2004. INCA: synonymous codon usage analysis and glycosylation. Biochemistry 50, 4441-4426. clustering by means of self-organizing map. Bioinformatics 20, 2329–2330. Larkin, M.A., Blackshields, G., Brown, N.P., Chenna, R., McGettigan, P.A., McWilliam, Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011. MEGA5: H., Valentin, F., Wallace, I.M., Wilm, A., Lopez, R., Thompson, J.D., Gibson, T.J., molecular evolutionary genetics analysis using maximum likelihood, Higgins, D.G., 2007. Clustal W and Clustal X version 2.0. Bioinformatics 23, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28, 2947–2948. 2731–2739. Lechner, J., Wieland, F., Sumper, M., 1985. Biosynthesis of sulfated saccharides N- Tomlinson, G.A., Jahnke, L.L., Hochstein, L.I., 1986. Halobacterium denitrificans sp. glycosidically linked to the protein via glucose. Purification and identification of nov., an extremely halophilic denitrifying bacterium. Int. J. Syst. Bacteriol. 36, sulfated dolichyl monophosphoryl tetrasaccharides from halobacteria. J. Biol. 66–70. Chem. 260, 860–866. VanDyke, D.J., Wu, J., Logan, S.M., Kelly, J.F., Mizuno, S., Aizawa, S., Jarrell, K.F., 2009. Leigh, J.A., Albers, S.V., Atomi, H., Allers, T., 2011. Model organisms for genetics in Identification of genes involved in the assembly and attachment of a novel the domain Archaea: methanogens, halophiles, Thermococcales and Sulfolobales. flagellin N-linked tetrasaccharide important for motility in the archaeon FEMS Microbiol. Rev. 35, 577–608. Methanococcus maripaludis. Mol. Microbiol. 72, 633–644. Liu, L.J., You, X.Y., Guo, X., Liu, S.J., Jiang, C.Y., 2011. Metallosphaera cuprina sp. nov., Vinogradov, E., Deschatelets, L., Lamoureux, M., Patel, G.B., Tremblay, T.L., an acidothermophilic, metal-mobilizing archaeon. Int. J. Syst. Evol. Microbiol. Robotham, A., Goneau, M.F., Cummings-Lorbetskie, C., Watson, D.C., Brisson, 61, 2395–2400. J.R., Kelly, J.F., Gilbert, M., 2012. Cell surface glycoproteins from Thermoplasma Lizak, C., Gerber, S., Numao, S., Aebi, M., Locher, K.P., 2011. X-ray structure of a acidophilum are modified with an N-linked glycan containing 6-C-sulfofucose. bacterial oligosaccharyltransferase. Nature 474, 350–355. Glycobiology 22, 1256–1267. Lynch, E.A., Langille, M.G., Darling, A.E., Wilbanks, E.G., Haltiner, C., Shao, K.S., Starr, Voisin, S., Houliston, R.S., Kelly, J., Brisson, J.R., Watson, D., Bardy, S.L., Jarrell, K.F., M.O., Teiling, C., Harkins, T.T., Edwards, R.A., Eisen, J.A., Facciotti, M.T., 2012. Logan, S.M., 2005. Identification and characterization of the unique N-linked Sequencing of seven haloarchaeal genomes reveals patterns of genomic flux. glycan common to the flagellins and S-layer glycoprotein of Methanococcus PLoS One 7, e41389. voltae. J. Biol. Chem. 280, 16586–16593. Magidovich, H., Eichler, J., 2009. Glycosyltransferases and Wieland, F., Heitzer, R., Schaefer, W., 1983. Asparaginylglucose: novel type of oligosaccharyltransferases in Archaea: putative components of the N- carbohydrate linkage. Proc. Natl. Acad. Sci. USA 80, 5470–5474. glycosylation pathway in the third domain of life. FEMS Microbiol. Lett. 300, Wright, F., 1990. The ‘effective number of codons’ used in a gene. Gene 87, 122–130. 23–29. Magidovich, H., Yurist-Doutsch, S., Konrad, Z., Ventura, V.V., Dell, A., Hitchen, P.G., Yan, Q., Lennarz, W.J., 2002. Studies on the function of oligosaccharyl transferase Eichler, J., 2010. AglP is a S-adenosyl-L-methionine-dependent subunits. Stt3p is directly involved in the glycosylation process. J. Biol. Chem. methyltransferase that participates in the N-glycosylation pathway of 277, 47692–47700. Haloferax volcanii. Mol. Microbiol. 76, 190–199. Yurist-Doutsch, S., Eichler, J., 2009. Manual annotation, transcriptional analysis, and Maita, N., Nyirenda, J., Igura, M., Kamishikiryo, J., Kohda, D., 2010. Comparative protein expression studies reveal novel genes in the agl cluster responsible for N structural biology of eubacterial and archaeal oligosaccharyltransferases. J. Biol. glycosylation in the halophilic archaeon Haloferax volcanii. J. Bacteriol. 191, Chem. 285, 4941–4950. 3068–3075. Matsumoto, S., Igura, M., Nyirenda, J., Matsumoto, M., Yuzawa, S., Noda, N., Inagaki, Yurist-Doutsch, S., Abu-Qarn, M., Battaglia, F., Morris, H.R., Hitchen, P.G., Dell, A., F., Kohda, D., 2012. Crystal structure of the C-terminal globular domain of Eichler, J., 2008. AglF, aglG and aglI, novel members of a gene island involved in oligosaccharyltransferase from Archaeoglobus fulgidus at 1.75 Å resolution. the N-glycosylation of the Haloferax volcanii S-layer glycoprotein. Mol. Biochemistry 51, 4157–4166. Microbiol. 69, 1234–1245. Mescher, M.F., Strominger, J.L., 1976. Purification and characterization of a Yurist-Doutsch, S., Magidovich, H., Ventura, V.V., Hitchen, P.G., Dell, A., Eichler, J., prokaryotic glucoprotein from the cell envelope of Halobacterium salinarium.J. 2010. N-glycosylation in Archaea: on the coordinated actions of Haloferax Biol. Chem. 251, 2005–2014. volcanii AglF and AglM. Mol. Microbiol. 75, 1047–1058. Meyer, B.H., Zolghadr, B., Peyfoon, E., Pabst, M., Panico, M., Morris, H.R., Haslam, Zähringer, U., Moll, H., Hettmann, T., Knirel, Y.A., Schäfer, G., 2000. Cytochrome S.M., Messner, P., Schäffer, C., Dell, A., Albers, S.V., 2011. Sulfoquinovose b558/566 from the archaeon Sulfolobus acidocaldarius has a unique Asn-linked synthase – an important enzyme in the N-glycosylation pathway of Sulfolobus highly branched hexasaccharide chain containing 6-sulfoquinovose. Eur. J. acidocaldarius. Mol. Microbiol. 82, 1150–1163. Biochem. 267, 4144–4149.

Homolog homolog homolog

4. Discussion

Originally discovered almost 40 years ago with the identification of an N-linked glycoprotein in the halophilic archaeon Halobacterium salinarum (Mescher and

Strominger, 1976), it has only been over the last decade that major progress in the understanding of archaeal N-glycosylation pathways has been made. Indeed, with more archaeal genome sequences and new molecular tools for working with the different species becoming available, we are likely to soon substantially expand our knowledge of this posttranslational modification in Archaea.

Haloferax volcanii, a halophilic archaeon originally isolated from the Dead Sea

(Mullakhanbhai and Larsen, 1975), was one of the first archeal species where the N- glycosylation pathway was addressed (Abu-Qarn and Eichler, 2006). The focus of my

Ph.D. research was to further describe the N-glycosylation process in Hfx. volcanii by identifying additional components of the pathway, as well as by biochemical characterization of enzymes known to be involved in the process.

4.1 AglJ, the first glycosyltransferase of the Hfx. volcanii Agl pathway

Mass spectrometry analysis of both the DolP glycan carrier and S-layer glycoprotein-derived peptides assigned AglJ as the GTase responsible for adding the first, as-yet-unidentified hexose to the pentasaccharide decorating the Hfx. volcanii S- layer glycoprotein. However, in contrast to the other GTases participating in Hfx. volcanii N-glycosylation (i.e., AglD, AglE, AglG, and AglI), where the deletion of the encoding gene led to the appearance of N-linked glycans totally lacking the sugar residue added by the GTase in question (Abu-Qarn et al., 2007; Abu-Qarn et al.,

2008; Yurist-Doutsch et al., 2008), small amounts of hexose-modified DolP and S-

layer glycoprotein-derived peptides were observed in cells lacking AglJ. This observation could be explained by a detailed analysis of the monosaccharide-charged

DolP pool in the parent and aglJ strains. Such analysis revealed the existence of three distinct monosaccharide-modified DolP carriers. The deletion of aglJ affected only one of these monosaccharide-modified lipid carriers. Thus, the other two minor peaks unaffected by the absence of AglJ were predicted to be generated by different

GTases. Indeed, one of these minor monosaccharide-charged DolP species was shown to be generated by HVO_1613. Moreover, in a later study, the GTase AglD was identified as the enzyme responsible for charging the third monosaccharide DolP

(Guan et al., 2010). In the case of the minor peak observed in the deletion strain at the position normally occupied by the AglJ-processed peak, it is conceivable that another

GTase inefficiently filled the void created by the absence of AglJ. Alternatively, this minor monosaccharide-modified DolP may be naturally present but is unmasked only in the absence of AglJ-catalyzed glycosylation of DolP. If so, then four distinct monosaccharide-modified DolP pools exist in Hfx. volcanii, namely those generated by the GTases AglJ, HVO_1613, AglD, and a fourth, currently unidentified GTase.

The results also confirmed that HVO_2601, previously identified as a Dpm1 homologue (Abu-Qarn and Eichler, 2006), does not serve such a role. It is also conceivable that AglJ does not act alone in adding the first hexose of the N-linked pentasaccharide, such that the relevant monosaccharide-modified lipid carrier and protein target observed in cells lacking AglJ reflect the residual contribution of a second protein involved in adding this first pentasaccharide sugar subunit.

Regardless of the agents responsible for generating the monosaccharide-charged

DolP species observed in the aglJ strain, no additional sugars are added to that minor monosaccharide-modified DolP carrier seen at the position normally occupied

by the AglJ-processed peak (nor, for that matter, to any of the other minor populations of monosaccharide-modified carriers). Likewise, no additional saccharides are bound to the monosaccharide N-linked to the S-layer glycoprotein in the aglJ strain. Thus, it seems unlikely that the monosaccharide added to the lipid carrier found at the same position as the AglJ-processed lipid carrier in the deletion strain represents an alternative linking sugar of a pentasaccharide variant decorating the S-layer glycoprotein. At the same time, the other minor monosaccharide-modified DolP generated through the actions of HVO_1613 does not appear to participate in generating the pentasaccharide N-linked to the S-layer glycoprotein. Moreover, given that the second and third sugars of the S-layer-linked pentasaccharide are hexuronic acids (Abu-Qarn et al., 2007), while the fourth sugar is a methyl ester of hexuronic acid (Magidovich et al., 2010), it is also unlikely that the hexose-charged DolP generated through HVO_1613 serves as a lipid carrier for those sugars ultimately found at pentasaccharide position 2, 3, or 4. As such, it appears that the two hexuronic acid and the methyl ester of hexuronic acid components of the N-linked pentasaccharide are derived from soluble activated species sequentially added onto an

AglJ-processed, monosaccharide-charged DolP carrier. The minor amount of monosaccharide-modified DolP carrier generated through the actions of HVO_1613 could participate in the biosynthesis of other glycoconjugates, such as glycolipids.

Indeed, the biogenesis of Hfx. volcanii glycolipids, a process of which little is known, was previously shown not to involve any of the Agl GTases (Naparstek et al., 2010).

Finally, AglD, previously shown to be involved in adding the final pentasaccharide subunit (Abu-Qarn et al., 2007), was subsequently found to be the GTase responsible for generation of the third monosaccharide-charged DolP identified in this study

(Guan et al., 2010).

In conclusion, AglJ was added to the growing list of Hfx. volcanii components shown to be involved in N-glycosylation in this haloarchaeon that includes AglB,

AglD, AglE, AglF, AglG, AglI, AglM, and AglP (Abu-Qarn et al., 2007; Abu-Qarn et al., 2008; Yurist-Doutsch et al., 2008; Magidovich et al., 2010; Yurist-Doutsch et al.,

2010), all of which (with the exception of AglD) are encoded by genes grouped into a single agl gene cluster (Yurist-Doutsch and Eichler, 2009).

4.2 The glycosyltransferase AglD

Having identified those GTases responsible for the addition of nucleotide activated sugars onto DolP carriers as part of the assembly of N-linked glycan, efforts next focused on the biochemical characterization of these enzymes. My attention was thus directed to AglD. Indeed, only limited experimental data on archaeal GTases involved in protein glycosylation were available at the time this study was undertaken.

My analysis of AglD represents a detailed examination of a GTase experimentally verified as participating in the modification of the Hfx. volcanii S- layer glycoprotein. Here, sequence alignment together with site-directed mutagenesis was used to identify AglD residues important for the catalytic action of the enzyme.

Specifically, AglD Asp110, Thr111 and Asp112 were shown to comprise the DXD motif typical of inverting GT-A fold-bearing GTases. In these enzymes, the DXD motif (or its variants) interacts with the Mg2+ or Mn2+ ion associated with a nucleotide-activated sugar donor (Lairson et al., 2008). Both AglD Asp110 and

Asp112 were shown to be essential for catalytic activity, although AglD D112E retained a limited amount of the native activity. In the case of Sinorhizobium meliloti

ExoM, where the DXD motif includes Asp96 and Asp98, it was shown that replacing

the former with an alanine completely eliminated enzymatic activity, whereas the same replacement at position 98 only led to a 70% loss of activity (Garinot-Schneider et al., 2000). In Salmonella enterica WbbE, where the DXD motif is expanded to include Asp93, Asp95 and Asp96, it was shown that exchanging Asp93 or Asp96 with an alanine abolished enzyme activity, while the same replacement at Asp95 only reduced that activity (Keenleyside et al., 2001). Thus, even within the GTase GT2 family (a major GTase family whose members include Hfx. volcanii AglD), differences in the organization and importance of DXD motif constituents exist.

AglD Asp201 was identified as the catalytic base of the enzyme. Just as the corresponding residue in S. enterica WbbE, i.e. Glu180, could be functionally replaced by an aspartic acid (Keenleyside et al., 2001), AglD D201E was also active.

By contrast, replacing the Asp187 catalytic base in S. meliloti ExoM led to a complete loss of function (Garinot-Schneider et al., 2000). This could indicate of differences in the donors and/or acceptors employed by each enzyme or point to unique mechanistic traits. In addition, although conserved in the archaeal AglD homologues examined in this study, Hfx. volcanii AglD Asp173 was not assigned as the catalytic base of the enzyme, given its functional replacement by a similarly-sized asparagine but not a similarly charged glutamic acid. Hence, it would appear that Asp173 is of structural, rather than catalytic importance to AglD activity.

In addition to these residues, AglD Arg139 and Arg152 also seem to be essential for enzyme function. Moreover, the AglD counterparts of other residues shown to be important for the catalytic activity of GT2 family members, such as S. meliloti ExoM Asp44 and Asp96 (Garinot-Schneider et al., 2000), may also play a role in the activity of the archaeal enzyme.

In conclusion, while the availability of three-dimensional structural information will elucidate the true contribution of these and other residues to AglD function, the in vivo assay I employed combined with sequence homology-based site-directed mutagenesis helped to determine the contribution of various AglD residues to the catalytic activity of this enzyme.

4.3 The archaeal flippase

Based on the physico-chemical properties of their phospholipids, cellular membranes present a hydrophobic barrier to the transfer of hydrophilic molecules.

Many biological processes, however, require that this barrier be overcome. To achieve this, membranes contain a varity of translocating proteins, including flippases, a class of proteins responsible for the delivery of lipid species across membranes (Sanyal and

Menon, 2009). Flippases are central components of N-glycosylation pathways across evolution, catalyzing the transfer of lipid-linked oligosaccharides (LLO) from one side of the membrane, where they are assembled, to the other side of the membrane, where they are translocated to their target proteins. While only limited information is available on N-glycosylation flippases in Eukarya and Bacteria, up until my studies nothing was known of flippase-mediated steps in the N-glycosylation pathway of

Archaea.

Bioinformatics approaches first identified AglR, a multi-spanning membrane protein, as a promising candidate for the role of archaeal flippase. These approaches were based on AglR homology to Wzx, a bacterial protein thought to translocate lipid-linked O-antigen precursor oligosaccharides across the plasma membrane. Both

AglR and Wzx contain a conserved domain that assigns them as members of the polysaccharide biosynthesis protein family (pfam01943). Members of this family are

integral membrane proteins often implicated in production of polysaccharides and include RfbX, a protein involved in the export of O-antigen and teichoic acid (Yao and Valvano, 1994; Liu et al., 1996; Raetz and Whitfield, 2002). Such homology, together with aglR being part of the agl gene cluster (Yurist-Doutsch and Eichler,

2009), assigned AglR as a possible flippase candidate.

I showed that Hfx. volcanii cells lacking AglR were able to accumulate DolP-

Man and lack the final mannose residue from the N-linked glycan decorating the S- layer glycoprotein both by mass spectrometry analysis and by the inability of the

AglR-lacking strain to incorporate tritium-radiolabeled mannose. At the same time, the accumulation of tetra- and trisaccharide-charged DolP species was also observed in the aglR strain, pointing to AglR serving alternate roles, such a, a more general flippase-related function. Alternatively, AglR could be required for the utilization but not the flipping of DolP-Man, as proposed for the eukaryal Lec35 protein (Anand et al., 2001). However, the fact that mono-, di-, tri- and tetrasaccharide-modified S-layer glycoprotein were detected in aglR cells but not the complete pentasaccharide suggests that AglR is indeed a DolP-Man flippase able to distinguish between DolP species bearing mannose and other hexoses. Such an assumption points to the involvement of multiple flippases in Hfx. volcanii N-glycosylation. The use of multiple flippases is also suggested to be the case for the eukaryal N-glycosylation pathway (Sanyal et al., 2008), although the identity of the eukaryal flippase(s) remains a matter of contention. Genetic approach proposed Rft1 to be the flippase delivering (GlcNAc)2-(Man)5-DolPP across the ER membrane in an ATP-independent manner (Helenius et al., 2002), however, later biochemical analysis put this into question (Frank et al., 2008; Rush et al., 2009). In the bacterium Campylobacter jejuni, PglK was assigned to act as an ATP-dependent flippase translocating UndPP-

heptasaccharide across the plasma membrane (Linton et al., 2005; Kelly et al., 2006).

However, this assignment can also be questioned since a pglK-lacking strain can still perform some N-glycosylation (Alaimo et al., 2006).

What other proteins could serve as additional Hfx. volcanii flippases or flippase- related components, responsible for delivery of the tetrasaccharide-charged DolP or its precursors across the plasma membrane? This role could be filled by an ABC transporter, acting along the lines of C. jejuni PglK. While the Hfx. volcanii genome includes 69 predicted genes encoding ABC transporters, corresponding to a broadly expanded repertoire relative to other sequenced haloarchaea and the second highest number of ABC transporters found in any available sequenced archaeal genome

(Hartman et al., 2010), a requirement for ATP in Hfx. volcanii LLO flipping has yet to be addressed.

While defining the precise function of AglR still requires an in vitro reconstitution of its activity, my results add up to the growing list of components of the N-glycosylation pathway in Hfx. volcanii and more specifically, to knowledge of

LLO flippases across evolution.

With the sum of the results obtained from my work and the work of others

(Abu-Qarn et al., 2007; Abu-Qarn et al., 2008; Yurist-Doutsch et al., 2008; Kaminski and Eichler, 2010; Yurist-Doutsch et al., 2010; Guan et al., 2010; Kaminski et al.,

2010; Kaminski et al., 2012; Cohen-Rosenzweig et al., 2012), a revised Hfx. volcanii glycosylation pathway can be drawn (Figure 8).

Figure 8: Schematic representation of the N-glycosylation pathway in Hfx. volcanii, circa 2013. The legend describes the components of the pentasaccharide decorating the S-layer glycoprotein.

4.4 New Agl pathway

In Hfx. volcanii and Hbt. salinarum, glycoproteins can be simultaneously modified by two distinct N-linked glycans, each attached via a different linking sugar

(Lechner and Wieland, 1989; Guan et al., 2012). Such complexity in N-glycosylation has not been reported beyond these two archaeal species. However, unlike Hbt. salinarum, where the pathway(s) of N-glycosylation has not yet been characterized, much is known of the parallel process in Hfx. volcanii. My research was directed at determining whether the same pathway used to generate the pentasaccharide N-linked to Hfx. volcanii S-layer glycoprotein Asn-13 and Asn-83 is also responsible for the assembly and attachment of the ‘low salt’ tetrasaccharide N-linked to Asn-498 when the cells are grown in ‘low salt’ medium (Guan et al., 2012).

Gene deletions, combined with mass spectrometric analysis of glycan-charged

DolP and the S-layer glycoprotein revealed that Agl proteins involved in the assembly

of the Asn-13- and Asn-83-linked pentasaccharide do not participate in the biosynthesis of the ‘low salt’ tetrasaccharide attached to Asn-498. Instead, the products of a distinct set of genes mediate the assembly and attachment of this glycan.

Given the substantial differences in the composition of the two N-linked glycans decorating the S-layer glycoprotein, it is not surprising that two distinct assembly pathways are involved. On the other hand, the finding that the OST AglB is not needed for ‘low salt’ tetrasaccharide attachment to S-layer glycoprotein Asn-498 is unexpected as AglB is essential for N-glycosylation of Asn-13 and Asn-83 in cells grown in either ‘high salt’ or ‘low salt’ medium. In Hfx. volcanii, and indeed in Hbt. salinarum, a single version of AglB is the only homologue of the eukaryal OST catalytic subunit, Stt3, or the bacterial OST, PglB, detected (Kaminski et al., 2013b).

As such, a currently unidentified OST is involved in the delivery of the ‘low salt’ tetrasaccharide (and its precursors) from DolP to S-layer glycoprotein Asn-498. The same may well hold true in Hbt. salinarum, where one N-linked glycan is transferred from a DolP carrier and the second from a DolPP carrier (Wieland et al., 1980;

Lechner and Wieland, 1989).

Based on the effects of deleting agl5-agl15 on DolP and S-layer glycoprotein glycosylation, a preliminary pathway for ‘low salt’ tetrasaccharide biogenesis can be drawn (Figure 9). In this working model, Agl5 and Agl6 are assigned roles in adding the linking hexose to DolP, while Agl7 contributes to the sulfation of this lipid-linked sugar. As such, the DolP-hexose seen in cells lacking Agl5 or Agl6 would correspond to this lipid charged with the first sugar of the pentasaccharide transferred to Asn-13 and Asn-83, a process that also occurs in low salt conditions (Guan et al., 2012).

Moreover, because cells lacking Agl7 contained DolP charged with a non-sulfated version of the ‘low salt’ tetrasaccharide while no Asn-498-fused ‘low salt’

tetrasaccharide (or its di- or trisaccharide precursors) were detected, sulfation of the

DolP-bound hexose may be needed for translocation of DolP charged with a more elaborate precursor ‘low salt’ tetrasaccharide or the complete glycan across the plasma membrane. Further work will be required to define the precise actions of

Agl5, Agl6 and Agl7, as well as the order in which they act. Agl8 and Agl9 contribute to the addition of a hexose to disaccharide-charged DolP. Similarly, Agl10-14 are involved in the appearance of the final rhamnose sugar of the ‘low salt’ tetrasaccharide on the DolP carrier. By contrast, cells lacking Agl15 assemble the intact ‘low salt’ tetrasaccharide on DolP, yet no such glycan is observed on S-layer glycoprotein Asn-498. This effect is consistent with Agl15 acting as a flippase, mediating the translocation of ‘low salt’ tetrasaccharide-charged DolP across the membrane. Accordingly, Agl15 shares 28% identity and 51% similarity with AglR which serves as the flippase of DolP-mannose during the assembly of the pentasaccharide added to S-layer glycoprotein Asn-13 and Asn-83 (Kaminski et al.,

2012).

Figure 9: Schematic representation of the second N-glycosylation pathway in Hfx. volcanii. The legend describes the components of the tetrasaccharide attached to S-layer glycoprotein Asn-498.

Presently, one can only speculate on why the Hfx. volcanii S-layer glycoprotein is modified by the two distinct N-linked glycans in low salt conditions but not at elevated salinity. While RNA encoding Agl5-Agl15 was detected in cells grown at both salt levels, quantitative PCR-based studies will be necessary to determine whether comparable levels of these mRNAs are present at the different salinities. A salt concentration-related conformational change in the S-layer glycoprotein leading to exposure of Asn-498 to the ‘low salt’ tetrasaccharide N-glycosylation machinery only at the lower salinity could be imagined. The fact that this position is not modified by the Asn-13/Asn-83-linked pentasaccharide when the cells are grown at higher salinity supports this hypothesis. However, the observation that barely detectable amounts of ‘low salt’ tetrasaccharide-bound DolP are seen in high salt conditions argues that modification of Asn-498 is a question of the availability of this glycan for transfer to this S-layer glycoprotein residue. Why cells lacking different components of the N-linked pentasaccharide biosynthetic pathway decorate Asn-498 with the ‘low salt’ tetrasaccharide at elevated salinity also awaits explanation.

In conclusion, this study identified components of a second Hfx. volcanii N- glycosylation pathway involved in the post-translational modification of the S-layer glycoprotein. Indeed, with the exception of Hbt. salinarum S-layer glycoprotein, such dual N-glycosylation of the same protein has not been reported elsewhere. While it has been shown that the variant surface glycoprotein of the eukaryotic parasite

Trypanosoma brucei can present two distinct N-linked glycans, the composition of the two oligosaccharides are very similar ((GlcNAc)2-(Man)5 and (GlcNAc)2-(Man)9), both relying on the same linking sugars. Moreover, both are added by isoforms of Stt3

(Izquierdo et al., 2009; Izquierdo et al., 2012). Thus, the finding that AglB is not the

OST of the novel Hfx. volcanii N-glycosylation pathway is striking. As such, an

enzyme not belonging to the Stt3/PglB/AglB family of OST is apparently capable of catalyzing the transfer of a lipid-linked glycan to a protein target. Future efforts will need to identify this protein, as well as to address other questions, including those related to the regulation of this novel N-glycosylation pathway, as well as its interplay with the previously described Hfx. volcanii N-glycosylation pathway involved in adding a pentasaccharide to select Asn residues of the S-layer glycoprotein.

4.5 N-glycosylation across the Archaea

The observation that while two N-glycosylation pathways exist in Hfx. volcanii but only one OST is identified made me consider the N-glycosylation pathways across evolution based on the identity of the OST. Since the first non-eukaryal N- glycosylated protein of Hbt. salinarum was reported in 1976 (Mescher and

Strominger, 1976), advances in describing both the structures of glycans N-linked to archaeal glycoproteins and archaeal N-glycosylation pathways were made over the next fifteen years. More recently, the availability of complete genome sequences, together with the development of appropriate molecular tools and techniques led to renewed interest in this topic. In the last decade, considerable progress has been made in addressing genes and proteins involved in N-glycosylation in several species.

Today, alongside such efforts being conducted at the molecular level, insight into the archaeal version of this universal post-translational modification can now be gleaned at the genome level.

Virtually all Archaea encode for components of a N-glycosylation pathway, pointing to such protein processing as being a common event in this life form.

Moreover, despite possessing a limited repertoire of GTases-encoding genes and with only relatively few experimentally confirmed examples, it is already abundantly clear

that archaeal N-glycosylation involves more variety in terms of sugars, glycan structures, and by extension, biosynthetic pathways than seen elsewhere. By focusing on the OST AglB as a single component of the archaeal N-glycosylation pathway, I have shown that gene duplication and modification occurred at numerous times during evolution. Furthermore, comparison of the organization and content of N- glycosylation genes in five members of the Haloferax genus revealed that substantial lateral gene transfer had occurred over the course of time.

Despite advances made in deciphering pathways of archaeal N-glycosylation, numerous unanswered questions remain. For instance, one can ask what species- specific changes allow the archaeal OST, AglB, to accommodate such a wide range of glycan structures. Do Archaea encountering similar environmental extremes decorate their proteins with similar N-linked glycans? How common is the ability to modify N- glycosylation in response to changing surroundings, a phenomenon recently observed in Hfx. volcanii? As new species appeared, did N-glycosylation change at the same rate? Finally, one can ask whether it will become possible to describe the composition of the N-linked glycans decorating archaeal glycoproteins based on their glycosylation gene content. Continued examination of archaeal N-glycosylation from the genomic perspective will help address these and elucidate other facets of the archaeal version of this universal protein-processing event.

In conclusion, during my research in the Eichler group, I helped to increase our understanding of N-glycosylation in Haloferax volcanii by increasing the number of known N-glycosylation pathway participants, together with the identification of a second, novel N-glycosylation pathway in this species. In addition, my computational analysis shed light on the N-glycosylation pathways across evolution.

Appendix A

"Distinct glycan-charged phosphodolichol carriers are required for

the assembly of the pentasaccharide N-linked to the Haloferax

volcanii S-layer glycoprotein"

Ziqiang Guan, Shai Naparstek, Lina Kaminski, Zvia Konrad and Jerry Eichler Mol. Microbiol. 2010. 78:1294-1303

Molecular Microbiology (2010) 78(5), 1294–1303 ᭿ doi:10.1111/j.1365-2958.2010.07405.x First published online 8 October 2010 Distinct glycan-charged phosphodolichol carriers are required for the assembly of the pentasaccharide N-linked to

the Haloferax volcanii S-layer glycoproteinmmi_7405 1294..1303

Ziqiang Guan,1† Shai Naparstek,2† Lina Kaminski,2 Introduction Zvia Konrad2 and Jerry Eichler2* 1Department of Biochemistry, Duke University Medical N-glycosylation is a post-translational modification per- Center, Durham, NC 27710, USA. formed by all three domains of life, namely Eukarya, Bac- 2Department of Life Sciences, Ben Gurion University, teria and Archaea (Helenius and Aebi, 2004; Eichler and Beersheva 84105, Israel. Adams, 2005; Szymanski and Wren, 2005; Weerapana and Imperiali, 2006). Presently, the pathway of N- glycosylation is best understood in higher Eukarya, where Summary the oligosaccharide covalently linked to select Asn residues of a target protein is first assembled from seven In Archaea, dolichol phosphates have been implicated soluble nucleoside-activated sugars sequentially added to as glycan carriers in the N-glycosylation pathway, a dolichol pyrophosphate carrier on the cytoplasmic much like their eukaryal counterparts. To clarify this face of the endoplasmic reticulum (ER) membrane. The relation, highly sensitive liquid chromatography/mass charged lipid carrier is then flipped to face the ER lumen, spectrometry was employed to detect and character- at which point seven additional sugars, derived from indi- ize glycan-charged phosphodolichols in the haloar- vidually charged and flipped phosphodolichol carriers, are chaeon Haloferax volcanii. It is reported that Hfx. added. Once assembled, the 14-meric oligosaccharide is volcanii contains a series of C and C dolichol phos- 55 60 transferred to the protein target (Burda and Aebi, 1999; phates presenting saturated isoprene subunits at the Helenius and Aebi, 2004). a and w positions and sequentially modified with the In contrast to the detailed delineation of the eukaryal first, second, third and methylated fourth sugar sub- N-glycosylation pathway, much less is known of this post- units comprising the first four subunits of the pen- translational modification in Archaea. As in the ER mem- tasaccharide N-linked to the S-layer glycoprotein, a brane, glycan-charged phosphodolichol species have reporter of N-glycosylation. Moreover, when this been detected in the archaeal plasma membrane. More- glycan-charged phosphodolichol pool was examined over, evidence exists assigning glycan-charged phospho- in cells deleted of agl genes encoding glycosyltrans- dolichols roles in archaeal N-glycosylation, in analogy to ferases participating in N-glycosylation and previ- the function these lipids serve in the parallel eukaryal ously assigned roles in adding pentasaccharide pathway (Burda and Aebi, 1999; Helenius and Aebi, residues one to four, the composition of the lipid- 2004). For example, the identical methylated hexasac- linked glycans was perturbed in the identical manner charide moiety as attached to the Methanothermus fervi- as was S-layer glycoprotein N-glycosylation in these dus surface (S)-layer glycoprotein is found on a dolichol mutants. In contrast, the fifth sugar of the pentasac- pyrophosphate carrier in this species (Hartmann and charide, identified as mannose in this study, is added Konig, 1989; Kärcher et al., 1993). Likewise, the sul- to a distinct dolichol phosphate carrier. This repre- phated polysaccharide moiety N-linked to the Halobacte- sents the first evidence that in Archaea, as in Eukarya, rium salinarum S-layer glycoprotein and flagellin was the oligosaccharides N-linked to glycoproteins are detected on dolichol phosphate intermediates, while C sequentially assembled from glycans originating from 60 dolichol phosphate species bearing glucose, mannose distinct phosphodolichol carriers. and N-acetylglucosamine units have also been observed in this organism (Mescher et al., 1976; Wieland et al., 1980; 1985; Lechner et al., 1985a; Sumper, 1987). Pulse- chase radiolabelling of Hbt. salinarum cells revealed the Accepted 16 September, 2010. *For correspondence. E-mail gradual transfer of the radiolabel from a lipid precursor to [email protected]; Tel. (+972) 8646 1343; Fax (+972) 8647 9175. †Equal contribution. the S-layer glycoprotein (Wieland et al., 1980), while the

© 2010 Blackwell Publishing Ltd Hfx. volcanii glycan-charged phosphodolichols 1295 incorporation of radiolabelled glucose into Haloferax vol- ryltransferase, responsible for delivery of the glycan to the canii glycoproteins was shown to proceed through a S-layer glycoprotein (Abu-Qarn and Eichler, 2006; Abu- glucose-containing phosphopolyisoprenol intermediate Qarn et al., 2007). (Zhu et al., 1995). Indeed, Hfx. volcanii membranes were Now, to address the involvement of glycan-charged reported to contain C55 and C60 dolichol phosphate phosphodolichols in the biosynthesis of the Hfx. volcanii charged with an a-linked mannosyl-(b1-4)-galactosyl N-linked pentasaccharide, liquid chromatography/mass group, and, to lesser extents, with a sulphated or phos- spectrometry (LC/MS) was employed to define the sugar phorylated dihexose and with a tetrasaccharide (compris- profiles of phosphodolichol-linked glycan carriers in a Hfx. ing the hexoses, mannose and galactose and the volcanii parent strain as well as from cells deleted of deoxyhexose, rhamnose), as well as with monosaccha- different agl genes. It is reported that Hfx. volcanii con- rides at radiochemical levels (Kuntz et al., 1997). Not all of tains a series of dolichol phosphate molecules sequen- these phosphodolichol-charged glycans have, however, tially modified with the one, two, three and four sugar been detected on Hfx. volcanii glycoproteins (Sumper subunits corresponding to the first four subunits of the et al., 1990; Mengele and Sumper, 1992; Abu-Qarn et al., pentasaccharide found on the S-layer glycoprotein. The 2007; Magidovich et al., 2010). fifth pentasaccharide subunit, mannose, is, in contrast, Thus, while glycan-charged phosphodolichols are impli- derived from its own phosphodolichol carrier. These find- cated in archaeal N-glycosylation, numerous questions ings thus not only provide the first direct evidence for the remain unanswered. Are those glycans found on dolichol sequential assembly of an oligosaccharide on a dolichol phosphate carriers in Archaea, presumably destined to carrier prior to the addition of that glycan to an archaeal decorate target proteins, assembled from soluble, acti- protein but also that the assembly of N-linked oligosac- vated monosaccharides, from monosaccharides trans- charides in Archaea involves glycans originating from dis- ferred from individual dolichol phosphate carriers or from tinct dolichol carriers, as occurs in Eukarya. both, as in Eukarya? If so, what is the relative contribution of each monosaccharide population in generating oligosaccharide-charged phosphodolichols? Is the Results protein-targeted oligosaccharide fully pre-assembled on a Hfx. volcanii contains a population of C and C single phosphodolichol carrier in the archaeal cytoplasm 55 60 dolichol phosphate molecules or does assembly of the phosphodolichol-charged oligosaccharide involve steps that transpire on both sides of In the present study, a total Hfx. volcanii lipid extract was the membrane? Finally, one can also ask whether assem- subjected to normal-phase liquid chromatography coupled bly of the N-linked glycan includes the addition of sugar with mass spectrometry. Figure 1A shows the total ion subunits to a glycan already transferred from its phospho- chromatogram of the NPLC-ESI/MS performed in the dolichol carrier to the protein target. Based on advances negative ion mode. The mass spectrum averaged from in describing the archaeal pathway of N-glycosylation those acquired during the retention time of 20–21 min made in the last 5 years (for review, see Yurist-Doutsch (Fig. 1B) shows prominent ion peaks of m/z 849.695 (this et al., 2008a; Calo et al., 2010), it may now be possible to and all reported values are for the monoisotopic ion peaks, address these and related questions. unless otherwise stated) and m/z 917.757, corresponding - In Hfx. volcanii, a series of Agl (archaeal glycosylation) to the [M–H] ions of the C55 and C60 dolichol phosphates proteins has been shown to participate in the assembly with two saturated isoprene units respectively. These mea- and attachment of a pentasaccharide decorating select sured ion masses are in agreement with the calculated - Asn residues of the S-layer glycoprotein, a reporter of values of m/z 849.690 for the [M–H] ion of the C55 dolichol - N-glycoslation in this species (Abu-Qarn and Eichler, phosphate and m/z 917.752 for the [M–H] ion of the C60 2006; Abu-Qarn et al., 2007; 2008; Yurist-Doutsch et al., dolichol phosphate. In addition, a very minor peak corre-

2008b; 2010; Kaminski et al., 2010; Magidovich et al., sponding to C50 dolichol phosphate (m/z 781.671) was 2010). The involvement of each of these proteins in the observed. Tandem mass spectrometry (MS/MS) was per- - N-glycosylation process was demonstrated by examining formed on the [M–H] ion at m/z 917.7 of C60 dolichol the N-linked glycan proﬁle of the S-layer glycoprotein phosphate (Fig. 1C); the obtained fragmentation pattern is isolated from Hfx. volcanii strains deleted of each of these consistent with the chemical structure previously agl genes, relative to the parent strain. As such, AglJ, described (Kuntz et al., 1997), with the saturated isoprene AglG, AglI, AglE and AglD were shown to participate in the units at both the a and the w positions. The same saturation introduction of the ﬁve sugar subunits comprising the pattern also held true for C55 dolichol phosphate (not

S-layer glycoprotein-bound pentasaccharide (Abu-Qarn shown). This is in contrast to the C55 undecaprenol involved et al., 2007; 2008; Yurist-Doutsch et al., 2008b; Kaminski in N-glycosylation in Bacteria, where the a position is et al., 2010) while AglB was shown to be the oligosaccha- unsaturated and the longer dolichols involved in eukaryal

Fig. 1. Normal-phase LC/MS identification of dolichol phosphate from the total lipid extract of Hfx. volcanii. A. Total ion chromatogram of the NPLC/MS analysis in the negative ion mode. - B. The [M–H] ions of C55 and C60 dolichol phosphate detected at m/z 849.695 and 917.757, indicated by C55 and C60 respectively. The mass spectrum shown is averaged from spectra acquired during the 20–21 min window, indicated by the shaded area in (A). - C. MS/MS of the [M–H] ion of C60 dolichol phosphate. The inset shows the predicted chemical structure of dolichol phosphate (according to Kuntz et al., 1997) and the MS/MS fragmentation scheme. The arrows indicating ¥20 and ¥50 reflect magnification of the ion peaks in the corresponding region of the m/z values on the graph.

N-glycosylation (C70–C110), where the a position is satu- chride was verified by MS/MS (Fig. 2D, right panel). The rated (Jones et al., 2009). The presence of two saturated most prominent ion at m/z 207 in the MS/MS spectrum isoprene units in archaeal dolichol phosphate is quite derived from the [M–2H]2- ion at m/z 810.46 is derived remarkable, considering that Bacteria contain only unsat- from the terminal (i.e. the fourth) sugar. Its methanol-less urated polyprenol phosphate, while in eukaryal dolichols, (32 Da) ion, shown at m/z 175, provides additional evi- only the a-isoprenes are saturated (Burda andAebi, 1999). dence for this sugar subunit being a methyl ester of hexu- Recently, the long-sought reductase for converting poly- ronic acid. prenol to dolichol in eukaryotic cells has been identified Finally, although the S-layer glycoprotein is ultimat- (Cantagrel et al., 2010). At present, no archaeal polyprenol ely modified by a N-linked pentasaccharide, no reductase has been described. pentasaccharide-modified dolichol phosphate species was detected. Instead, only dolichol phosphate species sequentially charged with the first four saccharides com- Apart from the complete pentasaccharide, Hfx. volcanii prising the pentasaccharide N-linked to the Hfx. volcanii contains dolichol phosphates charged with the same S-layer glycoprotein were observed. glycan series as found N-linked to the S-layer glycoprotein Hfx. volcanii cells lacking components of the Earlier studies revealed the Hfx. volcanii S-layer glycopro- N-glycosylation machinery present dolichol phosphates tein to be modified by an N-linked pentasaccharide com- void of or bearing truncated glycans prising a hexose, two hexuronic acids, a methyl ester of hexuronic acid and a final hexose (Abu-Qarn et al., 2007; To assess whether the various glycan-charged phospho- Magidovich et al., 2010). In addition, S-layer glycoprotein- dolichols described in the previous section are involved in derived peptides have also been shown to be modified by the N-glycosylation of the S-layer glycoprotein, the doli- glycans comprising the first, the first two, the first three chol phosphate-derived species from Hfx. volcanii cells and the first four sugar subunits of the N-linked pentasac- deleted of aglG, aglI, aglE and aglD were considered. charide (Abu-Qarn et al., 2007; 2008; Yurist-Doutsch Previous efforts implicated the products of these genes, et al., 2008b; 2010; Magidovich et al., 2010). To assess predicted glycosyltransferases, in the respective addition whether similar glycans also decorate Hfx. volcanii doli- of the second, third, fourth and fifth saccharides of the chol phosphate, phosphodolichol-linked glycans were pentasaccharide decorating the S-layer glycoprotein, profiled by NPLC/MS in the total lipid extract of Hfx. although direct biochemical proof for such activity has yet volcanii. Figure 2 (left panels) reveals the presence of to be provided (Abu-Qarn et al., 2007; 2008; Yurist- dolichol phosphate species modified by a glycan compris- Doutsch et al., 2008b). ing one to four saccharides, while MS/MS analysis con- When the total lipid extract from cells lacking AglG was

firmed the nature of the sugars added to the C60 dolichol analysed by NPLC/MS/MS as above, only hexose- phosphate (Fig. 2, right panels). Specifically, the fraction modified C55 and C60 phosphodolichols were detected eluting during the retention time of 16–16.5 min contains (Fig. 3). Likewise, cells deleted of aglI or aglE presented

C55 and C60 dolichol phosphate species modified by a dolichol phosphate species containing mono- and disac- hexose (peaks at m/z 1011.724 and 1079.797 respec- charides and mono-, di- and trisaccharides respectively. tively; Fig. 2A). In addition, a major peak at the m/z Consistent with these results, a recent study (Kaminski 1055.714, corresponding to a previously described sul- et al., 2010) showed that in the absence of AglJ, involved phoglycolipid (Sprott et al., 2003), was also observed. The in adding the first sugar subunit of the S-layer glycoprotein fraction eluting during the retention time of 20.8–21.3 min N-linked pentasaccharide, the level of a hexose-modified contains C55 and C60 dolichol phosphate species modified phosphodolichol species was significantly decreased, by a hexose and a hexuronic acid (peaks at m/z 1187.793 relative to what is seen in the parent strain. and 1255.858 respectively; Fig. 2B), while the fraction When dolichol phosphate-derived species from cells eluting during the retention time of 26–27 min contains C55 lacking AglD, implicated in adding the fifth and final and C60 dolichol phosphate species modified by a hexose subunit of the S-layer glycoprotein N-linked pentasaccha- and two hexuronic acids (peaks at m/z 1363.832 and ride, were compared with those of the parent strain, a very 1431.895 respectively; Fig. 2C). Finally, the fraction different effect was seen than observed in cells lacking eluting during the retention time of 35.5–36 min contains AglG, AglI or AglE. Hfx. volcanii membranes include three

C55 and C60 dolichol phosphate species modified by a hexose-modified phosphodolichol species retained at hexose, two hexuronic acids and a methyl ester of hexu- 14.77, 16.06 and 16.92 min. Of these, only the hexose- ronic acid (their doubly charged ions [M–2H]2- are modified phosphodolichol with a retention time of observed at m/z 766.43 and 810.46 respectively; Fig. 2D). 16.92 min was eliminated in the aglD deletion strain

The structure of the C60 phosphodolichol-linked tetrasac- (Fig. 4). This implies that AglD is a dolichol phosphate

Fig. 2. Normal-phase LC/MS/MS identification of mono-, di-, tri- and tetrasacchride-charged dolichol phosphate from the total Hfx. volcanii lipid extract. The left panels show the [M–H]- ions of (A) hexose-modified, (B) hexuronic acid-hexose-modified and (C) 2- dihexuronic-hexose-modified C55 and C60 phosphodolichol. The left panel of (D) shows the doubly charged [M–2H] ions of methyl ester of hexuronic acid-dihexuronic acid-hexose-modified C55 and C60 phosphodolichol, detected at m/z 766.43 and 810.46 respectively. RT refers to retention time. The right panels show the MS/MS spectra of the [M–H]- ion of (A) hexose-modified, (B) hexuronic acid-hexose-modified and 2- (C) dihexuronic acid-hexose-modified C60 phosphodolichol. The right panel of (D) shows the MS/MS spectrum of doubly charged [M–2H] ions of methyl ester of hexuronic acid-dihexuronic acid-hexose-modified C60 phosphodolichol. The inset in each right panel shows the chemical - 2- structure of the glycan-charged C60 phosphodolichol and the MS/MS fragmentation scheme of the [M–H] ion (or the [M–2H] ions in D). hexose synthase that catalyses the addition of a hexose subunits of the pentasaccharide N-linked to the Hfx. vol- residue to the lipid carrier. To identify the monosaccharide canii S-layer glycoprotein sequentially attach their respec- apparently added to the dolichol phosphate carrier by tive sugar substrates to a common dolichol phosphate

AglD, a mannose-charged C55 phosphodolichol standard carrier. In contrast, AglD, previously implicated in adding was examined as above. The mannose-modified phos- the fifth saccharide (now identified as mannose) to the phodolichol peak eluted at 16.99 min, just as was that N-linked glycan decorating the S-layer glycoprotein, adds peak missing from cells lacking AglD. Thus, the fifth and its substrate to a distinct dolichol phosphate carrier. final pentasaccharide subunit, added to its own dolichol phosphate carrier and previously reported to be a hexose Discussion (Abu-Qarn et al., 2007), is now identified as mannose. In conclusion, it is proposed that those enzymes previ- It has been long known that Archaea contain glycan- ously assigned roles in adding the first four saccharide bearing phosphodolichols and, like their eukaryal counter-

Fig. 3. LC/MS profiling of glycan-charged phosphodolichols in the parent and agl mutant strains. The presence or absence of mono-, di-, tri- and tetrasacchride-modified phosphodolichols in each strain was revealed by generating extracted ion chromatograms. The [M–H]- ion at m/z 1079.8, the [M–2H]2- ion at m/z 627.4, the [M–3H]3- ion at m/z 476.6 and the [M–3H]3- ion at m/z 540.0 were selected for monitoring the mono-, di-, tri- and tetrasacchride-modified phosphodolichols. Each of these four ions represents the highest abundance charge state observed by ESI/MS of the individual glycan-modified C60 phosphodolichol species. Above each peak, schematic representation of the linked glycan is shown. The full circles correspond to hexose, the full squares correspond to hexuronic acid and the CH3-bearing full square corresponds to the methyl ester of hexuronic acid. Note that the monosaccharide-modified phosphodolichol pool comprises several species.

parts (Burda and Aebi, 1999), roles for these lipids in Kuntz et al. (1997) ﬁrst reported the presence of C55

N-glycosylation were postulated (Mescher et al., 1976; and C60 dolichol phosphates saturated at both the a and Mescher and Strominger, 1978; Wieland et al., 1980; the w positions in Hfx. volcanii, including glycan-modified 1985; Lechner et al., 1985a,b; Sumper, 1987). Accord- species. The Hfx. volcanii dolichol phosphate pool was ingly, dolichol phosphates charged with either the identical reportedly modified by mannosyl-galactosyl groups, and or slightly modified versions of the glycans decorating to a lesser extent, by sulphated or phosphorylated dihexo- glycoproteins in Hbt. salinarum and M. fervidus were syl moieties and by a tetrasaccharide that includes reported (cf. Lechner and Wieland, 1989). Since little or mannose, galactose and rhamnose, sugars not detected nothing is known of the N-glycosylation process in these as components of the N-linked glycans reported to deco- species, much related to the involvement of dolichol phos- rate the S-layer glycoprotein at the time (Sumper et al., phates in N-glycosylation in Archaea remained a matter of 1990; Mengele and Sumper, 1992). However, given the speculation. The recent identification of a series of agl revision of the originally reported composition of the genes involved in the N-glycosylation of the Hfx. volcanii glycan N-linked to the Hfx. volcanii S-layer glycoprotein S-layer glycoprotein (for review, see Yurist-Doutsch et al., from a string of linear glucose residues (as well as a 2008a; Calo et al., 2010), however, now makes it possible glucose-, idose- and galactose-containing polysaccha- to address the precise role of dolichol phosphates as ride) (Sumper et al., 1990; Mengele and Sumper, 1992) to putative glycan carriers in the archaeal version of this a pentasaccharide comprising two hexoses, two hexu- post-translational modification. ronic acids and a methyl ester of hexuronic acid (Abu-

Fig. 4. The absence of AglD eliminates mannose-modified phosphodolichol. Normal-phase LC extracted ion chromatograms (EIC) of the dolichylphosphate-hexose [M–H]- ion at m/z 1079.8 from the parent strain (upper panel) and the DaglD (middle panel) are shown. The peaks at different retention times reflect the existence of three different dolichylphosphate-hexose species (Kaminski et al., 2010). The 16.92 min peak is eliminated in the mutant, as compared with the parent strain, suggesting AglD to be specific for the formation of the third monosaccharide-modified phosphodolichol species. The monosaccharide-modified phosphodolichol peak affected by the absence of AglD is retained at the position of a mannose-modified phosphodolichol standard (16.99 min; lower panel). The identities of the two other monosaccharide-modified phosphodolichols with retention times of 14.77 and 16.06 min, respectively, remain to be determined. While the results shown address C55 dolichol phosphate, similar results were obtained with C60 dolichol phosphate (not shown).

Qarn et al., 2007; Magidovich et al., 2010), the present tasaccharide subunit, as being soluble enzymes (Magi- study revisited the composition of glycans decorating doli- dovich et al., 2010; Yurist-Doutsch et al., 2010). chol phosphates in Hfx. volcanii in an attempt to link these In contrast to the sequential assembly of the first four glycan-charged lipids to the N-glycosylation process. pentasaccharide subunits onto a common dolichol phos- In recent work from our group (Kaminski et al., 2010), phate, the fifth subunit of the pentasaccharide, mannose, AglJ was shown to add the first hexose subunit of the was detected on its own distinct lipid carrier. The finding N-linked S-layer glycoprotein pentasaccharide to a doli- that AglD, involved in the addition of the fifth pentasaccha- chol phosphate carrier. In the present report, it was ride subunit, acts in a manner seemingly independent of revealed that the next three subunits of the pentasaccha- the other Agl proteins involved in generating the oligosac- ride are sequentially added to that AglJ-generated charide decorating the Hfx. volcanii S-layer glycoprotein is monosaccharide-charged carrier, through the respective not unexpected, given that aglD is the only gene not found actions of AglG, AglI and AglE. Since no hexuronic acid- in the agl gene island present in the Hfx. volcanii genome charged phosphodolichol was detected, it seems that (Yurist-Doutsch and Eichler, 2009). Thus, the observation pentasaccharide subunits two and three are added from that the same sugar subunits are found on both dolichol soluble, activated species. Likewise, the methyl ester of phosphate carriers and the S-layer glycoprotein, four of hexuronic acid found at position four of the N-linked pen- which are sequentially added to the lipid carrier in the same tasaccharide is added to the existing trisaccharide- order as found on the modified protein, together with the charged phosphodolichol from a soluble methylated fact that deletion of agl genes compromised dolichol phos- hexuronic acid species, since neither a dolichol phosphate glycosylation in a manner reminiscent of the effects phate modified with only a methyl ester of hexuronic acid of the same gene deletions on Hfx. volcanii S-layer glyco- nor a tetrasaccharide-charged phosphodolichol bearing a protein N-glycosylation (Abu-Qarn et al., 2007; 2008; hexuronic acid at position four was detected. These Yurist-Doutsch et al., 2008b), directly links dolichol phos- observations, moreover, offer support to the earlier phates, acting as mono- and oligosaccharide carriers, to assignment of the nucleoside-hexose dehydrogenase, the archaeal N-glycosylation process (Fig. 5). AglM, shown to catalyse the in vitro conversion of UDP- The confirmed involvement of archaeal dolichol phos- glucose to UDP-glucuronic acid and likely involved in the phate sugar carriers in Hfx. volcanii N-glycosylation pro- biogenesis of pentasaccharide subunits two, three and vides novel insight into the mechanism of this post- four, and of AglP, the SAM-dependent methyltransferase translational modification. Earlier work had shown that Agl responsible for modifying the fourth subunit of the pen- proteins involved in adding sugars found on the pentasac-

Fig. 5. The working model of the Hfx. volcanii N-glycosylation pathway. Select Asn residues of the Hfx. volcanii S-layer glycoprotein are modified by a pentasaccharide comprising a hexose, two hexuronic acids, a methyl ester of hexuronic acid and a terminal mannose residue. Based on the findings of the present study and earlier reports (Abu-Qarn et al., 2007; 2008; Plavner and Eichler, 2008; Yurist-Doutsch et al., 2008a; 2010; Kaminski et al., 2010; Magidovich et al., 2010), a working model of the Hfx. volcanii N-glycosylation pathway is provided. AglJ, AglG, AglI, AglE and AglD are assigned roles in either modifying dolichol phosphates or adding sugars to dolichol phosphate-bound sugars. AglB serves as the oligosaccharyltransferase, while AglF, AglP and AglM serve various sugar processing roles. At present, the flippase(s) responsible for delivering the lipid-charged glycans across the plasma membrane remain to be defined and are indicated by question marks. dolP, dolichol phosphate; NDP, nucleoside diphosphate.

charide N-linked to the Hfx. volcanii S-layer glycoprotein the fourth pentasaccharide subunit fails to occur (Magi- are membrane proteins oriented towards the cytoplasm, dovich et al., 2010). On the other hand, the actions of pointing to dolichol phosphate sugar charging as occur- AglP are not essential for modification of the S-layer gly- ring within the confines of the cell (Plavner and Eichler, coprotein by the tetrasaccharide formed in the absence of 2008). Hence, as no pentasaccharide-charged phospho- this methyltransferase. dolichol species could be detected, it is possible that In conclusion, despite considerable progress made in transfer of pentasaccharide subunit five occurs directly understanding archaeal N-glycosylation in recent years onto the S-layer glycoprotein-linked tetrasaccharide. (Yurist-Doutsch et al., 2008a; Calo et al., 2010), many Indeed, tetrasaccharide-modified S-layer glycoprotein- questions still remain unanswered. For instance, why derived peptides have been observed (Abu-Qarn et al., does N-glycosylation in some Archaea, such as Hfx. vol- 2007; Magidovich et al., 2010). Alternatively, transfer of canii, rely on dolichol phosphate while other species rely the fifth sugar subunit from its own dolichol phosphate on dolichol pyrophosphate or both, as in the case of Hbt. carrier to the lipid-linked tetrasaccharide and subsequent salinarum (Lechner and Wieland, 1989)? What is/are the transfer to the S-layer glycoprotein may occur too rapidly flippase(s) involved in N-glycosylation in Hfx. volcanii? to be detected here. This explanation for our inability to Finally, do the glycan-modified phosphodolichol species detect a Hfx. volcanii pentasaccharide-modified phospho- originally reported by Kuntz et al. (1997) participate in any dolichol species is unlikely, since such an entity was Hfx. volcanii post-translation modification? Continued readily observed upon examination of the dolichol phos- examination of the Hfx. volcanii N-glycosylation pathway phate pool of another halophilic archaea originating from will likely provide answers to these and other outstanding the Dead Sea, namely Haloarcula marismortui (Z. Guan, questions. D. Calo and J. Eichler, in preparation). To determine whether the fifth and final subunit of the pentasaccharide is added to the tetrasaccharide-charged phosphodolichol Experimental procedures to yield a potentially short-lived pentasaccharide-charged Strains and growth conditions lipid carrier or directly to the tetrasaccharide-modified The Hfx. volcanii parent strain WR536 (H53) and the same S-layer glycoprotein, additional biochemical studies are strain deleted of aglG, aglI, aglE or aglD were grown in required. It is, however, clear that methylation of the fourth complete medium containing 3.4 M NaCl, 0.15 M pentasaccharide subunit is important for addition of pen- MgSO4·7H2O, 1 mM MnCl2, 4 mM KCl, 3 mM CaCl2, 0.3% tasaccharide subunit five, since no N-linked pentasaccha- (w/v) yeast extract, 0.5% (w/v) tryptone, 50 mM Tris-HCl, pH ride is detected in the DaglP mutant, where methylation of 7.2, at 40°C (Mevarech and Werczberger, 1985). The prepa-

© 2010 Blackwell Publishing Ltd, Molecular Microbiology, 78, 1294–1303 1302 Z. Guan et al. ᭿ ration of Hfx. volcanii strains deleted of aglG, aglI, aglE and Biochemistry of the Duke University Medical Center and Z.G. aglD was previously reported (Abu-Qarn and Eichler, 2006; are supported by the LIPID MAPS Large Scale Collaborative Abu-Qarn et al., 2008; Yurist-Doutsch et al., 2008a). Grant No. GM-069338 from NIH. L.K. is the recipient of a Negev-Zin Associates Scholarship.

Isolation of the Hfx. volcanii lipid fraction References The total lipid contents of the Hfx. volcanii parent strain and of Hfx. volcanii DaglG, DaglI, DaglE and DaglD cells were Abu-Qarn, M., and Eichler, J. (2006) Protein N-glycosylation extracted as follows. Cells were harvested (8000 g, 30 min, in Archaea: defining Haloferax volcanii genes involved in 4°C) and frozen at -20°C until extraction was performed. At S-layer glycoprotein glycosylation. Mol Microbiol 61: 511– that point, the pelleted cells (15 g) were thawed, resus- 525. pended in 20 ml of double-distilled water (DDW) and DNase Abu-Qarn, M., Yurist-Doutsch, S., Giordano, A., Trauner, A., (1.7 mgml-1; Sigma, St. Louis, MO) and stirred overnight at Morris, H.R., Hitchen, P., et al. (2007) Haloferax volcanii room temperature. Methanol and chloroform were added to AglB and AglD are involved in N-glycosylation of the the cell extract to yield a methanol : chloroform : cell extract S-layer glycoprotein and proper assembly of the surface ratio of 2:1:0.8. After stirring for 24 h at room temperature, the layer. J Mol Biol 14: 1224–1236. mixture was centrifuged (1075 g, 30 min, 4°C). The clarified Abu-Qarn, M., Giordano, A., Battaglia, F., Trauner, A., Morris, supernatants were collected, combined and filtered through H.R., Hitchen, P., et al. (2008) Identification of AglE, a glass wool. Chloroform and DDW were added to the filtrate to second glycosyltransferase involved in N-glycosylation of yield a chloroform : DDW : filtrate ratio of 1:1:3.8, in a sepa- the Haloferax volcanii S-layer glycoprotein. J Bacteriol 190: rating funnel. After separation, the lower clear organic phase, 3140–3146. containing the total lipid extract, was collected into a round- Burda, P., and Aebi, M. (1999) The dolichol pathway of bottom flask and evaporated in a rotary evaporator at 35°C. N-linked glycosylation. Biochim Biophys Acta 1426: 239– For analysis of the dolichol phosphate-derived species, the 257. total lipid extracts were subjected to normal-phase LC/MS Calo, D., Kaminski, L., and Eichler, J. (2010) Protein glyco- analysis without pre-fractionation. sylation in Archaea: sweet and extreme. Glycobiology 20: 1065–1079. Cantagrel, V., Lefeber, D.J., Ng, B.G., Guan, Z., Silhavy, J.L., LC/MS and MS/MS Bielas, S.L., et al. (2010) SRD5A3 is required for converting polyprenol to dolichol and is mutated in a congenital Normal-phase LC-ESI/MS of lipids was performed using an glycosylation disorder. Cell 142: 1–15. Agilent 1200 Quaternary LC system coupled to a QSTAR XL Eichler, J., and Adams, M.W.W. (2005) Posttranslational quadrupole time-of-flight tandem mass spectrometer protein modification in Archaea. Microbiol Mol Biol Rev 69: (Applied Biosystems, Foster City, CA). An Ascentis Si HPLC 393–425. column (5 mm, 25 cm ¥ 2.1 mm) was used. Mobile phase A Hartmann, E., and Konig, H. (1989) Uridine and dolichyl consisted of chloroform/methanol/aqueous ammonium diphosphate activated oligosaccharides are intermediates hydroxide (800:195:5, v/v/v). Mobile phase B consisted of in the biosynthesis of the S-layer glycoprotein of Methano- chloroform/methanol/water/aqueous ammonium hydroxide thermus fervidus. Arch Microbiol 151: 274–281. (600:340:50:5, v/v/v/v). Mobile phase C consisted of Helenius, A., and Aebi, M. (2004) Roles of N-linked glycans in chloroform/methanol/water/aqueous ammonium hydroxide the endoplasmic reticulum. Annu Rev Biochem 73: 1019– (450:450:95:5, v/v/v/v). The elution programme consisted of 1049. the following: 100% mobile phase A was held isocratically for Jones, M.B., Rosenberg, J.N., Betenbaugh, M.J., and Krag, 2 min and then linearly increased to 100% mobile phase B S.S. (2009) Structure and synthesis of polyisoprenoids over 14 min and held at 100% B for 11 min. The LC gradient used in N-glycosylation across the three domains of life. was then changed to 100% mobile phase C over 3 min and Biochim Biophys Acta 1790: 485–494. held at 100% C for 3 min, and finally returned to 100% A over Kaminski, L., Abu-Qarn, M., Guan, Z., Naparstek, S., 0.5 min and held at 100% A for 5 min. The total LC flow rate Ventura, V.V., Raetz, C.R.H., et al. (2010) AglJ adds the was 300 ml min-1. The post-column splitter diverted ~10% of first sugar of the N-linked pentasaccharide decorating the LC flow to the ESI source of the Q-Star XL mass spec- the Haloferax volcanii S-layer glycoprotein. J Bacteriol trometer, with MS settings as follows: IS =-4500 V, doi:10.1128/JB.00705-10. CUR = 20 psi, GS1 = 20 psi, DP =-55 V and FP =-150 V. Kärcher, U., Schröder, H., Haslinger, E., Allmaier, G., For MS/MS, collision-induced dissociation (CID) was per- Schreiner, R., Wieland, F., et al. (1993) Primary structure of formed with collision energy ranging from 40 V to 70 V (labo- the heterosaccharide of the surface glycoprotein of Metha- ratory frame of energy) and with nitrogen as the collision gas. nothermus fervidus. J Biol Chem 268: 26821–26826. Data acquisition and analysis were performed using the Kuntz, C., Sonnenbichler, J., Sonnenbichler, I., Sumper, M., instrument’s Analyst QS software. and Zeitler, R. (1997) Isolation and characterization of dolichol-linked oligosaccharides from Haloferax volcanii. Acknowledgements Glycobiology 7: 897–904. Lechner, J., and Wieland, F. (1989) Structure and biosynthe- J.E. is supported by the Israel Science Foundation (Grant sis of prokaryotic glycoproteins. Annu Rev Biochem 58: 30/07). The mass spectrometry facility in the Department of 173–194.

Lechner, J., Wieland, F., and Sumper, M. (1985a) Biosynthe- Primary structure and glycosylation of the S-layer protein of sis of sulfated saccharides N-glycosidically linked to the Haloferax volcanii. J Bacteriol 172: 7111–7118. protein via glucose. Purification and identification of sul- Szymanski, C.M., and Wren, B.W. (2005) Protein glycosyla- fated dolichyl monophosphoryl tetrasaccharides from tion in bacterial mucosal pathogens. Nat Rev Microbiol 3: halobacteria. J Biol Chem 260: 860–866. 225–237. Lechner, J., Wieland, F., and Sumper, M. (1985b) Transient Weerapana, E., and Imperiali, B. (2006) Asparagine-linked methylation of dolichyl oligosaccharides is an obligatory protein glycosylation: from eukaryotic to prokaryotic step in halobacterial sulfated glycoprotein biosynthesis. systems. Glycobiology 16: 91R–101R. J Biol Chem 260: 8984–8989. Wieland, F., Dompert, W., Bernhardt, G., and Sumper, M. Magidovich, H., Yurist-Doutsch, S., Konrad, Z., Ventura, V.V., (1980) Halobacterial glycoprotein saccharides contain Hitchen, P.G., Dell, A., and Eichler, J. (2010) AglP is a covalently linked sulphate. FEBS Lett 120: 110–114. S-adenosyl-L-methionine-dependent methyltransferase Wieland, F., Paul, G., and Sumper, M. (1985) Halobacterial that participates in the N-glycosylation pathway of Halof- flagellins are sulfated glycoproteins. J Biol Chem 260: erax volcanii. Mol Microbiol 76: 190–199. 15180–15185. Mengele, R., and Sumper, M. (1992) Drastic differences in Yurist-Doutsch, S., and Eichler, J. (2009) Manual annotation, glycosylation of related S-layer glycoproteins from moder- transcriptional analysis and protein expression studies ate and extreme halophiles. J Biol Chem 267: 8182– reveal novel genes in the agl cluster responsible for 8185. N-glycosylation in the halophilic archaeon Haloferax Mescher, M.F., and Strominger, J.L. (1978) Glycosylation of volcanii. J Bacteriol 191: 3068–3075. the surface glycoprotein of Halobacterium salinarium via a Yurist-Doutsch, S., Chaban, B., VanDyke, D., Jarrell, K.F., cyclic pathway of lipid-linked intermediates. FEBS Lett 89: and Eichler, J. (2008a) Sweet to the extreme: protein gly- 37–41. cosylation in Archaea. Mol Microbiol 68: 1079–1084. Mescher, M.F., Hansen, U., and Strominger, J.L. (1976) For- Yurist-Doutsch, S., Abu-Qarn, M., Battaglia, F., Morris, H.R., mation of lipid-linked sugar compounds in Halobacterium Hitchen, P.G., Dell, A., and Eichler, J. (2008b) aglF, aglG salinarium. Presumed intermediates in glycoprotein and aglI, novel members of a gene cluster involved in synthesis. J Biol Chem 251: 7289–7294. the N-glycosylation of the Haloferax volcanii S-layer Mevarech, M., and Werczberger, R. (1985) Genetic glycoprotein. Mol Microbiol 69: 1234–1245. transfer in Halobacterium volcanii. J Bacteriol 162: 461– Yurist-Doutsch, S., Magidovich, H., Ventura, V.V., Hitchen, 462. P.G., Dell, A., and Eichler, J. (2010) N-glycosylation in Plavner, N., and Eichler, J. (2008) Defining the topology of Archaea: on the coordinated actions of Haloferax volcanii the N-glycosylation pathway in the halophilic archaeon AglF and AglM. Mol Microbiol 75: 1047–1058. Haloferax volcanii. J Bacteriol 190: 8045–8052. Zhu, B.C., Drake, R.R., Schweingruber, H., and Laine, R.A. Sprott, G.D., Larocque, S., Cadotte, N., Dicaire, C.J., McGee, (1995) Inhibition of glycosylation by amphomycin and M., and Brisson, J.R. (2003) Novel polar lipids of halophilic sugar nucleotide analogs PP36 and PP55 indicates that eubacterium Planococcus H8 and archaeon Haloferax Haloferax volcanii beta-glucosylates both glycoproteins volcanii. Biochim Biophys Acta 1633: 179–188. and glycolipids through lipid-linked sugar intermediates: Sumper, M. (1987) Halobacterial glycoprotein biosynthesis. evidence for three novel glycoproteins and a novel sulfated Biochim Biophys Acta 906: 69–79. dihexosyl-archaeol glycolipid. Arch Biochem Biophys 319: Sumper, M., Berg, E., Mengele, R., and Strobel, I. (1990) 355–364.

Appendix B

"Protein glycosylation in Archaea: Sweet and Extreme"

Doron Calo, Lina Kaminski and Jerry Eichler Glycobiology. 2010. 20:1065-1076

Glycobiology vol. 20 no. 9 pp. 1065–1076, 2010 doi:10.1093/glycob/cwq055 Advance Access publication on April 5, 2010 REVIEW Protein glycosylation in Archaea: Sweet and extreme Downloaded from Doron Calo2, Lina Kaminski2, and Jerry Eichler1,2 ings, such as those characterized by extremes in salinity, pH or 2Department of Life Sciences, Ben Gurion University of the Negev, Beersheva temperature, to live in sulfur-based environments or to produce 84105, Israel methane as a by-product of their anaerobic respiration (cf. Rothschild and Mancinelli 2001). However, with the subse- Received on March 7, 2010; revised on March 31, 2010; accepted on March 31, 2010 quent widespread application of 16S rRNA analysis, it http://glycob.oxfordjournals.org/ became clear that Archaea are major denizens of ‘normal’ bi- While each of the three domains of life on Earth possesses ological niches, including seawater, soil and even our own fl unique traits and relies on characteristic biological strate- intestinal ora (DeLong 1998; Chaban, Ng, et al. 2006). In- gies, some processes are common to Eukarya, Bacteria and deed, it is now recognized that Archaea are major players in Archaea. Once believed to be restricted to Eukarya, it is the ecosystem, important for processes as diverse as the Earth's now clear that Bacteria and Archaea are also capable of nitrogen cycle and global warming (Francis et al. 2007; Gal- performing N-glycosylation. However, in contrast to Bac- perin 2007). teria, where this posttranslational modification is still As befitting a distinct form of life, Archaea possess traits considered a rare event, numerous species of Archaea, iso- that distinguish them from either Bacteria or Eukarya, in addi- lated from a wide range of environments, have been tion to their characteristic rRNA. For example, archaeal at Ben-Gurion University of the Negev, Aranne Library on November 18, 2013 reported to contain proteins bearing Asn-linked glycan membranes are composed of phospholipids of strikingly differ- moieties. Analysis of the chemical composition of the ent composition than those used elsewhere. Whereas bacterial Asn-linked polysaccharides decorating archaeal proteins and eukaryal phospholipids comprise fatty acyl groups ester- has, moreover, revealed the use of a wider variety of sugar linked to the sn-1,2 positions of glycerol, in Archaea, mem- subunits than seen in either eukaryal or bacterial glycopro- brane lipids comprise polyisoprenyl groups ether-linked to teins. Still, although first reported some 30 years ago, little the sn-2,3 positions of a glycerol backbone (Sprott 1992). had been known of the steps or components involved in the On the other hand, Archaea share much in common with both archaeal version of this universal posttranslational modifi- Bacteria and Eukarya. Like Bacteria, Archaea are single-celled cation. Now, with the availability of sufficient numbers of organisms surrounded by a single membrane and lack internal genome sequences and the development of appropriate ex- organelles. Moreover, in both cases, the genome is organized perimental tools, molecular analysis of archaeal N- as a single, circular chromosome, often divided into operons glycosylation pathways has become possible. Accordingly (Koonin and Wolf 2008). When, however, replication, DNA using halophilic, methanogenic and thermophilic model packaging, transcription and other aspects of information pro- species, insight into the biosynthesis and attachment of cessing are considered, Archaea are more reminiscent of ! N-linked glycans decorating archaeal glycoproteins is Eukarya (Sandman and Reeve 2000; Bell and Jackson 2001). starting to amass. In this review, current understanding Thus, archaeal biology can be considered a mosaic of archaeal- of N-glycosylation in Archaea is described. specific, bacterial-like and eukaryal-like traits. As such, the study of Archaea has not only revealed biological strategies Keywords: Archaea/extremophiles/N-glycosylation/ not seen elsewhere but has also served to reveal links between posttranslational modification bacterial and eukaryal processes previously thought to be unrelated, as well as providing first examples of universal biological phenomena. With respect to protein glycosylation, Archaea Archaea: the third domain of life have provided examples of each of these three scenarios. The Archaea were first recognized as a distinct domain of life, unrelated to either Bacteria or Eukarya, in 1977, as a result of Archaeal N-linked glycoproteins: here, there and Carl Woese's pioneering use of 16S ribosomal (r)RNA analysis everywhere (Fox et al. 1977; Woese and Fox 1977). At the time, those microorganisms assigned to the archaeal domain were all Since Neuberger reported that a carbohydrate group was an in- extremophiles, i.e. species able to thrive in the face of some tegral part of ovalbumin some 75 years ago (Neuberger 1938), of the most physically challenging conditions on the planet. it was believed that N-glycosylation was a trait restricted to Eu- Archaea were shown to reside in seemingly ‘harsh’ surround- karya. This belief was, however, challenged in 1976 with the demonstration that the surface (S)-layer glycoprotein of the ex- 1To whom correspondence should be addressed: Tel: +972-8646-1343; Fax: treme halophile, Halobacterium salinarum, experiences this +972-8647-9175; e-mail: [email protected] same processing event (Mescher and Strominger 1976a). De-

Fig. 1. The composition of representative N-linked glycans decorating archaeal glycoproteins. N-Linked glycans from (A) the Hbt. salinarum S-layer glycoprotein, at Ben-Gurion University of the Negev, Aranne Library on November 18, 2013

(B) the M. fervidus S-layer glycoprotein and (C) S. acidocaldarius cytochrome b558/566 are presented. In (A), the upper glycan is found at Hbt. salinarum S-layer glycoprotein Asn-2, while the lower glycan is found at 10 other sequons. The abbreviations used here but not in the text are: fGal, galactofuranose; GalUA, galacturonic acid; GlcUA, glucuronic acid; Man, mannose; Me, methyl; Qui, quinovose. tailed analysis showed that the S-layer glycoprotein is modified 1975; Sumper et al. 1990). Since, N-linked glycans presenting by two different Asn-linked oligosaccharides, with Asn-2 being enormous diversity in their sugar composition have been iden- decorated with a repeating sulfated pentasaccharide, bound via tified in numerous archaeal proteins derived from a variety of an N-glycosylamine bond, and 10 other Asn residues being species isolated from a wide range of environmental niches. modified with a sulfated glycan, linked through a glucose res- Methanogenic Archaea, possessing the ability to generate idue (Wieland et al. 1980; Wieland et al. 1983; Lechner et al. CH4 from CO2 and H2 or other carbon sources and detected 1985a; Lechner and Wieland 1989)(Figure 1A). At the time of in environments spanning a broad spectrum of temperature, sa- first being shown to contain N-linked glycans, Hbt. salinarum linity, pressure and pH (Thauer et al. 2008), also express was defined as an obligate halophilic bacterial species and con- glycoproteins bearing N-linked glycans. For example, the sidered an odd member of the bacterial world both because of Methanothermus fervidus S-layer glycoprotein is modified by its unusual habitat and due to its ability to N-glycosylate protein a hexasaccharide composed of a N-acetylgalactosamine (Gal- targets. However, with the realization that life on Earth com- NAc) subunit linked to the target asparagine residue, three prised three distinct domains, i.e. Eukarya, Bacteria and mannose residues and finally, two methylated hexoses (man- Archaea (Woese and Fox 1977), Hbt. salinarum was reassigned nose or glucose) at the nonreducing end of the glycan to the archaeal branch of the universal tree of life, with its abil- (Figure 1B; Kärcher et al. 1993). In Methanococcus voltae,a ity to perform N-glycosylation now offering further support for trisaccharide composed of a N-linked N-acetylglucosamine the similarity of Archaea to Eukarya and their distinctiveness (GlcNAc), a 2,3-diacetamido-2,3-dideoxy-β-glucuronic acid from Bacteria, as suggested by the original 16S rRNA analysis group and a terminal 2-acetamido-2-deoxy-β-mannuronic acid used to distinguish between the three forms of life. Indeed, as moiety, with the carbonyl group at C-6 forming an amide bond discussed below, the concept of archaeal and eukaryal symme- with the amino group of threonine, was detected on both the S- try gained further support from early investigations into the layer glycoprotein and flagellins (Voisin et al. 2005). In some archaeal N-glycosylation pathway. strains, this trisaccharide is augmented by an extra 220 or Not long after the Hbt. salinarum S-layer glycoprotein was 262 Da entity of unknown identity (Chaban et al. 2009). Final- identified as the first noneukaryal N-modified glycoprotein, ly, the tetrameric glycan N-linked to Methanococcus other similarly modified archaeal polypeptides were described. maripaludis flagellin is reminiscent of its M. voltae counter- These included the Hbt. salinarum flagellin, shown to bear the part. In M. maripaludis, the glycan comprises a N-linked same glycan-linked sulfated polysaccharide as the S-layer gly- GalNAc, followed in turn by a 2,3-diacetamido-2,3-dideoxy- coprotein from this species (Wieland et al. 1985), and the S- β-glucuronic acid group, as in the M. voltae glycan, and a 3- layer glycoprotein of Haloferax volcanii, a halophilic species acetamidino derivative of 2,3-diamino-2,3-dideoxymannuronic first isolated from the Dead Sea (Mullakhanbhai and Larsen acid amidated with a threonine amino group found in the M. 1066 Archaeal N-glycosylation

Table I. N-glycosylation in the three domains of life

Eukarya Bacteria Archaea

Essential property? Yes No No Sugar donor Nucleotide-activated sugars Nucleotide-activated Nucleotide-activated sugars? Dolichol-phosphate-bound glycans sugars Dolichol-(pyro)phosphate-bound glycans? Oligosaccharide-charged lipid carrier Dolichol pyrophosphate Undecaprenol Dolichol pyrophosphate pyrophosphate Dolichol phosphate (Hbt. salinarum, Hfx. volcanii) Flippase Process ATP independent ATP independent Unknown Identity Rft1 may be involved in flipping PglK Unknown OST Downloaded from Oligomeric status Multimeric Single subunit Single subunit Catalytic subunit Stt3 PglB AglB Localization Lumenal face of the ER membrane External surface of cell External surface of cell Sequon recognized N-X-S/T; X≠P D/E-Z-N-X-S/T; Z,X≠P N-X-S/T; X≠P (N-X-N/L/V, Hbt. salinarum) OST type Type E Type B Types A, B and E Modification of N-linked glycan? Trimming in ER, elaboration in Golgi None reported None reported http://glycob.oxfordjournals.org/ voltae glycan at this position. The M. maripaludis glycan is, based motif reportedly employed in bacterial (i.e. Campylobac- however, capped by the unique terminal sugar, 2-acetamido- ter jejuni) N-glycosylation where X and Z are any residue but 2,4-dideoxy-5-O-methyl-hexos-5-ulo-1,5-pyranose. This ap- proline (Kowarik et al. 2006; Abu-Qarn and Eichler 2007). parently represents the first example of a naturally occurring Still, although relatively few sequons in archaeal glycoproteins diglycoside of an aldulose (Kelly et al. 2009). have been experimentally verified as being modified, the ami- Thermophilic and hyperthermophilic Archaea, the latter no acid composition within and surrounding these sequons can growing optimally at temperatures above 80°C, also contain be distinguished from the comparable positions in eukaryal fi fi experimentally con rmed N-modi ed glycoproteins. Thermo- glycoproteins (Abu-Qarn and Eichler 2007). For instance, if at Ben-Gurion University of the Negev, Aranne Library on November 18, 2013 plasma acidophilum, growing optimally at 56°C and pH 2, the modified Asn is considered as position 0, there is a high lacks a cell wall but is surrounded by a highly glycosylated probability of finding aromatic residues at positions −2and membrane protein reported to contain a branched glycan large- −1, a small hydrophobic residue (Gly or Val) at the X position ly based on mannose subunits yet also containing glucose and and a larger hydrophobic residue (Ile, Leu, Met, Phe, Trp or galactose moieties, asparagine-linked through GlcNAc (Yang Tyr) at the +1 position in a eukaryal glycoprotein (Ben-Dor and Haug 1979a). Cytochrome b558/566 from Sulfolobus acid- et al. 2004; Petrescu et al. 2004). In contrast, in only few ar- ocaldarius, originally isolated from solfatara regions in chaeal sequons confirmed as being modified are Phe, Trp or Yellowstone National Park and shown to grow optimally at Tyr detected at positions −2and−1, while Ser or Thr are read- 75 to 80°C and pH 2 to 3 (Brock et al. 1972), is modified ily found at the X position and Ala and Gly predominate at the by a hexasaccharide that includes a glucose subunit, two man- +1 position. No example in Archaea of an aromatic residues nose residues, a 6-sulfoquinovose subunit and two GlcNAc being present at the +1 position has been reported. Indeed, groups, one of which serves as the linking sugar the only time a Tyr was detected at this site was in a sequon (Figure 1C; Hettmann et al. 1998; Zähringer et al. 2000). experimentally shown to be nonmodified (Lechner and Sumper Although considered common in glycolipids of chloroplasts 1987). In vitro analysis of oligosaccharyltransferase (OST) ac- and photosynthetic Bacteria (Imhoff et al. 1982), 6-sulfoqui- tivity in the hyperthermophile, Pyrococcus furiosus, revealed novose (or 6-deoxy-6-sulfoglucose) had not been previously that modification of an Asn-X-Thr-bearing reporter peptide detected in a glycoprotein. was more efficient than was processing of the same peptide For a listing of additional confirmed or proposed N-modified presenting an Asn-X-Ser sequon, with either the sequon Thr archaeal glycoproteins, along with the evidence supporting the or Ser being essential for glycosylation and the presence of claim for glycan modification, the reader is directed to the re- Pro at the X position preventing sequon modification (Igura view by Eichler and Adams (2005). et al. 2008). On the other hand, as elaborated below in the section dealing with the archaeal OST, modifications in archaeal sequon composition are tolerated in Hbt. salinarum, suggesting Not like everybody else that predictive algorithms may have overlooked archaeal gly- Today, it is clear that organisms from all three domains are ca- coproteins bearing similarly modified or even novel N- pable of performing N-glycosylation (Weerapana and Imperiali glycosylation sites (Zeitler et al. 1998). 2006; Abu-Qarn, Eichler, et al. 2008). Still, archaeal N-glyco- The linking sugar serving to attach an oligosaccharide to a sylated proteins can be distinguished from their eukaryal and modified Asn residue also differs in glycoproteins across the bacterial counterparts not only by the extremophilic nature of three domains of life. In bacterial N-glycoproteins characterized their source species or the unusual sugars employed but also in thus far, bacillosamine (2,4-diamino-2,4,6-trideoxyglucopyra- terms of the amino acid sequence surrounding modified se- nose) serves as the linking sugar (Young et al. 2002), whereas quons and with respect to how oligosaccharides are N-linked in the vast majority of eukaryal N-glycoproteins, GlcNAc serves to target proteins (Table I). this role (Spiro 1973). Rare instances of eukaryal reliance on As in Eukarya, archaeal N-glycosylation occurs as NXS/T- other linking sugars have been reported, such as glucose in based sequons, in contrast to the more elaborate D/EZNXS/T- the case of laminin (Schreiner et al. 1994). In Archaea, a va- 1067 D Calo et al. riety of saccharides have been shown to serve as linking sug- observation that in Hbt. salinarum, the glycan moiety of the ar, including glucose, GlcNAc and GalNAc. Indeed, as noted pyrophosphodolichol-bound sulfated polysaccharide is also de- above, in the Section Archeal N-linked glycoproteins: here, tected on the S-layer glycoprotein and flagellins in this species there and everywhere, two different linking sugars can be (Lechner et al. 1985a; Wieland et al. 1985). The sulfated poly- employed within the same Archaeal glycoprotein, as exempli- saccharide is methylated in the pyrophosphodolichol-linked fied by the Hbt. salinarum S-layer glycoprotein. The ability form but not when protein-bound (Lechner et al. 1985b). In to use different linking sugars in the same glycoprotein is es- contrast, the hexasaccharide moiety attached to the M. fervidus pecially intriguing when one considers that Hbt. salinarum S-layer glycoprotein retains the methylation introduced at the seemingly only encodes a single OST (Magidovich and Eich- dolicholpyrophosphate carrier level (Hartmann and Konig ler 2009; see below). 1989; Kärcher et al. 1993), as also seems to be the case in

Hfx. volcanii, where methylation of the S-layer glycoprotein- Downloaded from bound pentasaccharide is observed (Magidovich et al. 2010). Early insight into the archaeal N-glycosylation pathway Encouraged by the conclusion made at the time that N-glyco- Facing the outside sylation was restricted to Eukarya and Archaea, initial efforts When the topology of N-glycosylation was considered, further fi aimed at de ning the pathway responsible for this posttransla- parallels between the eukaryal and archaeal pathways were http://glycob.oxfordjournals.org/ tional modification in Archaea sought further similarities shown, with evidence obtained pointing to archaeal N-glyco- between the archaeal pathway and its eukaryal counterpart. sylation as occurring on the outer surface of the cell, the topological equivalent of the lumenal-facing leaflet of the en- The sugar carrier doplasmic reticulum (ER) membrane bilayer, the site of N- As in Eukarya (Burda and Aebi 1999), the archaeal N-linked glycosylation in Eukarya. Localization of archaeal N-glycosyl- oligosaccharide is apparently assembled on a dolichol carrier, ation to the external cell surface is supported by studies rather than the undecaprenol carrier used in bacterial N-glyco- showing the ability of Hbt. salinarum cells to modify cell-im- sylation (Szymanski and Wren 2005). Accordingly, Archaea permeable, sequon-bearing hexapeptides with sulfated contain both dolichol phosphate and dolichol pyrophosphate oligosaccharides (Lechner et al. 1985a). Although unable to at Ben-Gurion University of the Negev, Aranne Library on November 18, 2013 bearing either mono- or polysaccharides. Specifically, Hbt. sal- cross the haloarchaeal plasma membrane (Mescher and Stro- minger 1978), bacitracin is nonetheless able to interfere with inarum has been reported as containing C60-dolichols carrying glucose, mannose and GlcNAc units, as well as a sulfated tet- N-glycosylation in Hbt. salinarum, ultimately preventing the rasaccharide (Mescher et al. 1976; Lechner et al. 1985a). Hfx. transfer of sulfated oligosaccharides to the S-layer glycoprotein volcanii contains mannosyl-(β1-4)-galactosyl phosphodoli- (Mescher and Strominger 1978; Wieland et al. 1980). Addi- chol, lesser quantities of sulfated or phosphorylated dihexosyl tional support for archaeal N-glycosylation occurring on the phosphodolichol and dolichol phosphate bearing a tetrasacchar- outer cell surface came with the description of membrane- ide comprising mannose, galactose and rhamnose subunits. In bound pyrophosphatases that orient their active site to the exterior (Meyer and Schäfer 1992; Amano et al. 1993). It was this haloarchaeon, the glycans were all linked to C55-and/or proposed that such enzymes could participate in the archaeal C60-dolichol moieties (Kuntz et al. 1997). The saccharide- charged dolichol species in Hfx. volcanii are unusual in that N-glycosylation process by dephosphorylating dolichol pyro- the ω-terminal isoprene unit is saturated and since only mono- phosphate, presumably following transfer of oligosaccharides phosphorylated dolichol is observed. from the lipid carrier to protein targets (Meyer and Schäfer Proof for the involvement of such glycan-charged dolichols 1992). Accordingly, an externally oriented membrane-bound in archaeal protein N-glycosylation has also been provided. pyrophosphatase from the thermoacidophile Sulfolobus toko- The transfer of radiolabeled glucose from uridine diphosphate daii has been recently shown to be able to hydrolyze (UDP)-[3H]glucose to glycoproteins was shown to proceed isopentenylpyrophosphate and geranylpyrophosphate (Manabe through a glucose-containing phosphopolyisoprenol interme- et al. 2009). Finally, studies supporting a cotranslational diate in Hfx. volcanii (Zhu et al. 1995). Treatment with mode of membrane protein insertion in Archaea also lend bacitracin, a compound that interferes with recycling of pyro- support to protein glycosylation transpiring on the exterior phosphodolichyl polysaccharide carriers following release of surface of the cell (Gropp et al. 1992; Dale and Krebs their bound oligosaccharides (Stone and Strominger 1971), 1999; Ring and Eichler 2004). was able to prevent modification of the Hbt. salinarum S-layer glycoprotein by the glucose-linked sulfated glycan otherwise Enzymes of glycosylation added to 10 sequons of the protein (Mescher and Strominger In addition to these early efforts aimed at delineating steps and 1978). In contrast, bacitracin treatment failed to prevent addi- components of the archaeal N-glycosylation pathway, several tion of the GlcNAc-linked repeating sulfated pentasaccharide groups also addressed enzymes putatively involved in this at Hbt. salinarum S-layer glycoprotein Asn-2, suggesting the posttranslational modification, including enzymes proposed use of dolichol phosphate rather than dolichol pyrophosphate to participate in the assembly of the oligosaccharide-charged as the carrier for this glycan (Wieland et al. 1980). Indeed, the phosphodolichol carrier. GlcNAc transferase activity was par- reportedly exclusive use of phosphodolichyl sugar carriers in tially characterized from Hbt. salinarum membranes (Mescher Hfx. volcanii N-glycosylation is reflected in the inability of et al. 1976). Photo-affinity experiments using 5-azido-[32P] bacitracin to prevent such protein modification in this species UDP-glucose identified a putative phosphodolichol glucose (Kuntz et al. 1997; Eichler 2001). Further evidence linking do- synthase in Hfx. volcanii homogenates (Zhu et al. 1995), while lichol-charged glycans to N-glycosylation comes with the phosphodolichol mannose synthase, susceptible to inhibition

1068 Archaeal N-glycosylation by amphomycin, was purified from T. acidophilum (Zhu and Laine 1996). Thus, the first two decades of research into archaeal N-glycosylation provided glimpses of selected aspects of the process, in a variety of model systems. Detailed understanding of the archaeal version of this posttranslational modification would have to, however, wait until the dawn of the genome era.

Homology shows the way

The examination of Archaea at the genome level began in Downloaded from 1996, when the ﬁrst complete sequence of an archaeon (corresponding to the fourth genome sequenced overall), namely that of the methanogen Methanocaldococcus (then Methanococcus) jannaschii, was published (Bult et al. 1996).

With the additional archaeal genome sequences now available http://glycob.oxfordjournals.org/ (today, over 140 archaeal genomes are at various stages of completion (http://genomesonline.org/index2.htm; Oct., 2009)), bioinformatics approaches were enlisted in the search for components of the archaeal N-glycosylation pathway. Anal- ysis of the genome of Archaeoglobus fulgidus,asulfur- metabolizing organism that grows optimally at 83°C (Klenk et al. 1997), first revealed the presence of gene clusters containing sequences predicted to encode elements of a N- glycosylation pathway (Burda and Aebi 1999). The major breakthrough in identifying components of archaeal N-glyco- at Ben-Gurion University of the Negev, Aranne Library on November 18, 2013 sylation pathways soon followed, coming from efforts combining the identification of archaeal homologues of known elements of the parallel eukaryal and bacterial pathways, together with gene deletion and subsequent analysis of the effects of such deletion on the N-glycosylation of reporter glycoproteins. In this manner, Abu-Qarn and Eichler (2006) and Chaban, Voisin, et al. (2006), respectively, studying N-glycosylation in Hfx. volcanii and M. voltae, simultaneously identified gene products experimentally verified as participating in this posttranslational modification.

Sweet and salty: N-glycosylation in Haloferax volcanii As the major polypeptide in the species and previously shown to be amenable to the study of other posttranslational modifications (e.g. signal peptide cleavage (Fine et al. 2006)and protein lipid modifi cation(Konrad and Eichler 2002)), the Hfx. volcanii S-layer glycoprotein represents an excellent model for addressing archaeal N-glycosylation. At the time of its description (Sumper et al. 1990; Mengele and Sumper 1992), it was reported that of the seven putative N-glycosylation sites present in the S-layer glycoprotein, Asn-13 and Asn-505 were each modified by a linear chain of β-1-4 linked glucose resi-

Fig. 2. Working models of archaeal N-glycosylation pathways. Currently identified components of the N-glycosylation pathways of (A) Hfx. volcanii, (B) M. voltae and (C) M. maripaludis. The abbreviations used are: dolP, dolichol phosphate; dolPP, dolichol pyrophosphate; NDP, nucleoside diphosphate; 1P, 1 phosphate. Dolichol, embedded in the schematically- depicted plasma membrane (drawn as the two parallel horizontal lines), is represented by a vertical line. The identities of the sugars represented by each shape are listed next to Asn-linked glycan in each panel. In (B), the optional introduction of a 220 or 260 Da sugar seen in some M. voltae species (Chaban et al. 2009)isreflected by the use of parentheses. 1069 D Calo et al. dues, while Asn-274 and/or Asn-279 were described as bearing N-glycosylation in the methanogens a glucose-, idose- and galactose-containing polysaccharide. Following the discovery that the same novel N-linked trisac- Later efforts relying on more sophisticated mass spectrometry charide is attached to both the S-layer glycoprotein and tools revealed the Hfx. volcanii S-layer glycoprotein to instead fl ff be modified at Asn-13 and Asn-83 by a pentasaccharide com- agellins in M. voltae (Voisin et al. 2005), e orts focused on defining the pathway of N-glycosylation in this obligate anaer- prising two hexoses, two hexuronic acids and a 190 Da species obic methanogen. The fact that the same N-linked glycan (Abu-Qarn et al. 2007), subsequently shown to correspond to a fl methyl ester of hexuronic acid (Magidovich et al. 2010). It was decorates both the S-layer glycoprotein and agellins in M. voltae (asisalsothecaseinHbt. salinarum (Wieland et al. also shown that the sequon at Asn-370 is not modified (Abu- 1985)) points to a common pathway for N-glycosylation in a Qarn et al. 2007). given species. As with Hfx. volcanii, studies of the N-glycosyl- In addition to redefining the N-linked glycan profile of the Downloaded from ation pathway of M. voltae began with the search for S-layer glycoprotein, these more recent studies also served to homologues of known eukaryal or bacterial N-glycosylation define components of the Hfx. volcanii N-glycosylation path- genes (Chaban, Voisin, et al. 2006). Accordingly, this strategy way. The initial use of bioinformatics tools to identity Hfx. led to the identification of aglA, shown to participate in the ad- volcanii homologues of eukaryal and bacterial N-glycosylation dition of the terminal sugar residue of the trisaccharide, namely genes (Abu-Qarn and Eichler 2006) led to the identification of N-acetyl mannuronic acid linked to threonine, as well as of the http://glycob.oxfordjournals.org/ open reading frames, subsequently remained agl (archaeal gly- OST, aglB. Through its ability to complement a conditionally cosylation) genes following the nomenclature of Chaban, lethal alg7 mutation in yeast, it was concluded that AglH is a Voisin, et al. (2006) and shown to encode for glycosyltrans- GlcNAc-1-phosphate transferase, catalyzing the transfer of ferases (GTs) (i.e. AglD, AglE, AglI and AglJ), an OST (i.e. UDP-activated GlcNAc to dolichol pyrophosphate and respon- AglB) and other sugar-processing enzymes (i.e. AglF) involved sible for the first step in the M. voltae N-glycosylation process in the assembly and attachment of the pentasaccharide decorat- (Shams-Eldin et al. 2008). Finally, the second sugar residue of ing the Hfx. volcanii S-layer glycoprotein (Abu-Qarn et al. 2007; the trisaccharide, a diacetylated glucuronic acid, is added by Abu-Qarn, Giordano, et al. 2008; Yurist-Doutsch et al. 2008). the combined actions of two enzymes, AglC and AglK (Cha- aglG, encoding a putative GT, was next identified through ban et al. 2009), as reflected in Figure 2B. it being positioned between the aglB and aglI sequences at Ben-Gurion University of the Negev, Aranne Library on November 18, 2013 The N-linked glycan decorating the flagellins of M. maripa- (Yurist-Doutsch et al. 2008). Indeed, when the various agl ludis corresponds to a tetrasaccharide in which the sugar sequences were mapped to the Hfx. volcanii genome, it was noted that apart from aglD, all were sequestered to a gene residues found at positions two and three are essentially identical to the two outer sugar subunits of the M. voltae N-linked island stretching from aglJ to aglB (Yurist-Doutsch and Eichler trisaccharide (Kelly et al. 2009). As such, M. maripaludis 2009). However, as the current annotation of the Hfx. volcanii AglA was identified as participating in the addition of the sugar genome did not recognize the aglE sequence as an open read- subunit at position three of the tetrasaccharide, an acetylated ing frame (Abu-Qarn, Giordano, et al. 2008), the agl gene and acetamidino-modified mannuronic acid linked to threonine island was subjected to manual reannotation. In this manner, ad- (VanDyke et al. 2009). However, given that the linking sugar ditional sequences, i.e. aglP, aglQ and aglR,wereidentified of the M. maripaludis glycan is GalNAc, rather than the (Yurist-Doutsch and Eichler 2009), while algM was shown to GlcNAc employed in M. voltae glycoproteins, the M. maripa- lie beyond the original gene island borders (Yurist-Doutsch ludis enzyme involved in adding the diacetylated glucuronic et al. 2010). acid to position two of the tetrasaccharide, AglO, is not homol- Although each of the identified Hfx. volcanii Agl proteins ogous to either M. voltae AglC or AglK (VanDyke et al. 2009). has been assigned a specific role through a combination of Finally, AglL was assigned as participating in either the attach- gene deletion and mass spectrometry approaches, only AglF, ment of the threonine to the third sugar residue or the addition AglM and AglP have been characterized biochemically. AglF of the terminal sugar, the novel 2-acetamido-2,4-dideoxy-5-O- was shown to be a glucose-1-phosphate uridyltransferase in- methyl-hexos-5-ulo-1,5-pyranose (VanDyke et al. 2009) volved in the biosynthesis of the hexuronic acid found at (Figure 2C). position three of the S-layer glycoprotein-linked pentasaccharide (Yurist-Doutsch et al. 2010). AglM, a UDP-glucose dehydrogenase, was shown to participate in the biogenesis of Pass the sugar: AglB, the archaeal OST the hexuronic acid found at pentasaccharide position two and likely of the hexuronic acids found at positions three and four, OSTs serve to transfer lipid-linked polysaccharides to select as well (Yurist-Doutsch et al. 2010). In a combined in vitro re- Asn residues of target proteins. While the OST of higher Eu- constitution experiment, AglF and AglM were shown to work karya exists as a multimeric complex based on the Stt3 subunit in a coordinated manner to generate UDP-glucuronic acid from (Zufferey et al. 1995), the OST of Archaea, like that of Bacte- glucose-1-phosphate and uridine triphosphate (UTP) in a ria, comprises only a single component, i.e. AglB (Abu-Qarn NAD+-dependent manner (Yurist-Doutsch et al. 2010). Finally, and Eichler 2006; Chaban, Voisin, et al. 2006; Igura et al. AglP was confirmed to be a S-adenosyl-L-methionine-depen- 2008). AglB was first identified in Hfx. volcanii and M. voltae, dent methyltransferase responsible for the formation of the when it was shown that deletion of the encoding gene led to a methyl ester of hexuronic acid found at position four of the loss of N-glycosylation of reporter glycoproteins (Abu-Qarn pentasaccharide (Magidovich et al. 2010). The current working and Eichler 2006; Chaban, Voisin, et al. 2006; Abu-Qarn et model of the Hfx. volcanii N-glycosylation pathway is pre- al. 2007). Like its eukaryal (i.e. Stt3) and bacterial (i.e. PglB) sented in Figure 2A. counterparts, AglB comprises a multiple membrane-spanning

1070 Archaeal N-glycosylation

Table II. All three types of OST catalytic centers are found in archaeal AglB proteins

Type A (WWDYG and DM motifs) Type B (WWDYG and MI motifs) Type E (WWDYG and DK motifs)

Euryarchaeota Euryarchaeota Crenarchaeota Archaeoglobus fulgidus Ferroplasma acidarmanusa Caldivirga maquilingensis Archaeoglobus profundus Methanobrevibacter ruminantium Desulfurococcus kamchatkensis Candidatus Methanoregula boonei Methanobrevibacter smithii Hyperthermus butylicus Candidatus Methanosphaerula palustris Methanocaldococcus fervens Ignicoccus hospitalis Ferroglobus placidus Methanocaldococcus infernus Metallosphaera sedula Haloarcula marismortui Methanocaldococcus jannaschii Pyrobaculum aerophilum Halobacterium salinarum R1 Methanocaldococcus SP FS406-22 Pyrobaculum arsenaticum

Halobacterium sp. NRC-1 Methanocaldococcus vulcanius Pyrobaculum calidifontis Downloaded from Haloferax volcanii Methanococcus aeolicus Pyrobaculum islandicum Halogeometricum borinquense Methanococcus maripaludis C5 Staphylothermus hellenicus Halomicrobium mukohataei Methanococcus maripaludis C6 Staphylothermus marinus Haloquadratum walsbyi Methanococcus maripaludis C7 Sulfolobus acidocaldarius Halorhabdus utahensis Methanococcus maripaludis S2 Sulfolobus islandicus Halorubrum lacusprofundi Methanococcus vannielii Sulfolobus solfataricus

Haloterrigena turkmenica Methanococcus voltae Sulfolobus tokodaii http://glycob.oxfordjournals.org/ Methanocella paludicola Methanosphaera stadtmanae Thermofilum pendens Methanococcoides burtonii Methanothermobacter thermautotrophicus Thermoproteus neutrophilus Methanocorpusculum labreanum Picrophilus torridusa Euryarchaeota Methanoculleus marisnigri Thermoplasma volcaniuma Pyrococcus abyssi Methanosaeta thermophila Thermoplasma acidophiluma Pyrococcus furiosus Methanosarcina acetivorans Pyrococcus horikoshii Methanosarcina barkeri Thermococcus barophilus Methanosarcina mazei Thermococcus gammatolerans Methanospirillum hungatei Thermococcus kodakarensis Natrialba magadi Thermococcus onnurineus Natronomonas pharaonis Thermococcus sibiricus Uncultured methanogenic archaeon RC-I Thermococcus sp. AM4 at Ben-Gurion University of the Negev, Aranne Library on November 18, 2013 Thaumarchaeota Nitrosopumilus maritimus Nanoarchaeota Nanoarchaeum equitans Korarchaeota Candidatus Korarchaeum cryptofilum aThe MI motif corresponds to LxxI/VxxxV in these species and is much farther from the WWDYG motif than in other AglB proteins.

N-terminal domain and a soluble C-terminal domain that in- structural information, reassessment of the alignment of Stt3/ cludes the WWDYG motif implicated in the catalytic PglB/AglB family member sequences revealed the presence mechanism of AglB/Stt3/PglB proteins (Wacker et al. 2002; of a spatially proximal DxxK motif, proposed to form the ac- Yan and Lennarz 2002; Igura et al. 2008). Also as observed with tive site of the enzyme along with the WWDYG motif of the eukaryal OSTs (Turco et al. 1977; Munoz et al. 1994)andC. protein. Indeed, the Asp and Lys residues of this DxxK motif jejuni PglB (Glover et al. 2005; Linton et al. 2005), AglB can were shown to be important for yeast Stt3 activity by site-di- also transfer truncated glycan structures (Chaban, Voisin, et al. rected mutagenesis. In the P. furiosus enzyme, the active site 2006; Abu-Qarn et al. 2007), pointing to the relaxed substrate motifs assume unusual conformations, with the first three resi- specificity of the archaeal enzyme. In addition, AglB is appar- dues of the WWDYG motif predicted to adopt a rare left- ently also able to transfer not only fully assembled handed helical conformation and the DxxK motif being part oligosaccharides but also precursor polysaccharides to target of an unusually long six-residue helix. proteins. Hfx. volcanii AglB is, moreover, unusual in that it The more recent solution of the C. jejuni PglB structure transfers polysaccharides from dolichol phosphate rather (Maita et al. 2010) revealed an impressive degree of folding than dolichol pyrophosphate carriers (Kuntz et al. 1997; similarity to P. furiosus AglB. However, such comparison also Eichler 2001; Abu-Qarn et al. 2007). showed that rather than being universal in Stt3/PglB/AglB Insight into OST action in Archaea, and indeed, across evo- family members, the P. furiosus AglB DxxK motif is replaced lution, has been provided by solution of the three-dimensional by a MxxI motif in the bacterial enzyme. Based on this obser- structure of the C-terminal soluble domain of P. furiosus AglB vation, a reassessment of sequence alignment data extended the (Igura et al. 2008). Providing the first portrayal of a catalytic P. furiosus AglB DxxK motif to DxxKxxx[MI], now termed domain of a Stt3/PglB/AglB family member at atomic level the DK motif. Likewise, the MxxI motif of C. jejuni PglB resolution, this study revealed the C-terminal soluble domain was extended to encompass a MxxIxxx[IVW] motif, now to comprise four structural regions, assembled into a novel ar- termed the MI motif. When the distribution of these motifs chitecture. The WWDYG motif, known to participate in Stt3 across Stt3/PglB/AglB family members in the three domains and PglB function, is found within a β-helix-based central core of life was addressed, the differential distribution of the MI region that also includes an 80 residue, antiparallel β-barrel- and DK motifs, as well as of a variant of the DK motif com- like element. The central core is surrounded by two peripheral prising a DxxMxxx[KI] signature (the DM motif), allowed domains, each mainly containing β-strands. Guided by this evolutionary patterns to be drawn. It was shown that together 1071 D Calo et al. with the WWDYG motif, eukaryal Stt3 proteins and archaeal family and the GT-B fold-containing retaining GT4 family as AglB proteins from members of the archaeal phyla Crenarch- being the prototypes that eventually gave rise to the array of gly- aeota (including P. furiosus) and some members of the phylum cosyltranferases seen today (Lairson et al. 2008). Furthermore, Euryarchaeota employ the DK motif to present an E-type cat- unlike what is observed in Eukarya and Bacteria, where known alytic center, as do the sole sequenced members of the archaeal and predicted GTs are respectively assigned to 67 and 54 differ- phyla, Korarchaeota, Nanoarchaeota and Thaumarchaeota. ent GT families, proteins identified as archaeal GTs are Along with bacterial PglB proteins, other members of Eur- distributed among only 13 GT families. yarchaeota, including M. voltae and M. maripaludis, employ In Hfx. volcanii, aglB is found within a gene cluster that the WWDYG and MI motifs to form the B-type catalytic cen- includes the agl GTs of this species, with the exception of ter. Yet other Euryarchaeota, such as Hfx. volcanii, combine the aglD, as well as non-GT-encoding agl sequences serving oth-

WWDYG and DM motifs to create the A-type catalytic center. er N-glycosylation-related roles (Yurist-Doutsch and Eichler Downloaded from Archaeal AglB proteins thus contain catalytic centers of all 2009; Magidovich et al. 2010; Yurist-Doutsch et al. 2010). three types (Table II). Moreover, in those archaeal species en- In contrast, the agl genes of M. maripaludis S2 are more coding multiple AglB sequences (cf. Magidovich and Eichler widely distributed in the genome, with only aglC being 2009), only a single type of catalytic center is found in all of found in proximity to aglB (VanDyke et al. 2009). In other the predicted AglB proteins of that species. Archaea, varying degrees of such aglB-centered clustering of Bioinformatics-based assignment of multiple versions of GT-encoding genes are seen (Magidovich and Eichler 2009). http://glycob.oxfordjournals.org/ AglB being present within a given archaeal species still re- While substantial gene clustering is observed in halophiles quires experimental confirmation that each sequence is in and many methanogens, this arrangement is seen 3-fold less fact expressed and capable of OST activity. Indeed, justifica- often in hyperthermophilic species. The evolutionary signifi- tion for the existence of multiple versions of AglB in a cance of this observation remains to be considered. single species is not obvious if one considers the seeming pro- In addition to the different verified and predicted N-glyco- miscuity of Hbt. salinarum AglB. In this species, it was shown sylation pathway components described in halophilic and that replacement of the Ser-4 residue of the S-layer glycopro- methanogenic archaea, additional archaeal sugar-modifying tein 2Asn-3Ala-4Ser sequon with Val, Leu or Asn did not enzymes have been reported. For instance, the pathway of prevent N-glycosylation at the Asn-2 position (Zeitler et al. UDP-acetamido sugar biosynthesis has been described for two at Ben-Gurion University of the Negev, Aranne Library on November 18, 2013 1998). Moreover, the same OST is apparently responsible for members of the class Methanococcales, namely M. maripaludis adding both the repeating sulfated pentasaccharide moiety and M. jannaschii. In each case, the pathway enzymes having through a GalNAc link to Asn-2 as well as the sulfated poly- been purified following expression in Escherichia coli and bio- saccharide unit attached via a glucose subunit at 10 other N- chemically characterized (Namboori and Graham 2008). glycosylation sites of the S-layer glycoprotein (Lechner and Specifically, M. maripaludis MMP1680 (or M. jannaschii Wieland 1989). MJ1420) catalyzes the isomerization and transamination of fructose-6-P to create α-D-glucosamine-1-phosphate, which is α Putative N-glycosylation pathway components in other converted to -D-glucosamine-1-phosphate by the M. maripalu- Archaea dis MMP1077 (or M. jannaschii MJ1100) phosphomutase. M. maripaludis MMP1076 (or M. jannaschii MJ1101) then Although pathways of N-glycosylation have only been out- catalyzes both the acetylation of glucosamine-1-phosphate lined in Hfx. volcanii, M. voltae and M. maripaludis,N- and the transfer of the resulting GlcNAc-1-phosphate to UTP glycosylated proteins have been identified in numerous archae- to generate UDP-GlcNAc and pyrophosphate. M. maripaludis al species found in a variety of environments (cf. Eichler and MMP0705 (or M. jannaschii MJ1504) can isomerize UDP- Adams 2005). In the vast majority of these species, nothing is GlcNAc to create UDP-N-acetyl-α-D-mannosamine (UDP- known of the N-glycosylation process. ManNAc), which is oxidized by M. maripaludis MMP0706 As a first step to redressing this situation, the Carbohydrate- (or M. jannaschii MJ0468) to produce UDP-N-acetylmannosa- Active Enzymes (CAZy) database (http://www.cazy.org/fam/ minuronate (UDP-ManNAcA). Given the homologies of acc_GT.html; Cantarel et al. 2009) was consulted to address MMP1680 (or MJ1420) to bacterial GlmS, of MMP1077 (or the presence and distribution of predicted GTs in 56 archaeal MJ1100) to GlmM and of MMP1076 (or MJ1101) to GlmU, genomes (Magidovich and Eichler 2009). In addition to iden- these studies point to methanoarchaea as producing acetamido tifying AglB in all but two species considered (i.e. Aeropyrum sugars, e.g. GlcNAc, using the bacterial pathway (Mengin- pernix and Methanopyrus kandleri), it was also shown that Lecreulx and van Heijenoort 1994), rather than the Leloir path- GTs assigned to the GT2 and GT4 families predominate in way used by Eukarya (Milewski et al. 2006). Still, despite Archaea (Magidovich and Eichler 2009), as reported elsewhere the fact that acetamido sugars have been detected on the (Lairson et al. 2008). The fewest GT2 and GT4 glycosyltrans- N-linked glycans of methanoarchaeal glycoproteins (see section ferases are detected in hyperthermophilic Archaea. In fact, the on N-glycosylation in methanogens and Figure 1), the involve- observation that Nanoarchaeum equitans, a hyperthermophile ment of these enzymes in N-glycosylation remains to be encoding the smallest genome known (490,885 base pairs; Hu- demonstrated. The same is true of a P. furiosus sugar nucleoti- ber et al. 2002; Waters et al. 2003), contains just a single GT2 dyltransferase with extremely broad sugar and nucleotide and two GT4 glycosyltransferases, while Ignicoccus hospitalis substrate specificity (Mizanur et al. 2004) as well as of phos- (the symbiotic host of N. equitans; Paper et al. 2007) encodes phohexomutases from Pyrococcus horikoshii (Akutsu et al. only two GT2 and a single GT4 glycosyltransferases has been 2005), Sulfolobus solfataricus (Ray et al. 2005)andThermo- suggested to reflect the GT-A fold-containing inverting GT2 coccus kodakaraensis (Rashid et al. 2004), reported to possess

1072 Archaeal N-glycosylation both phosphoglucomutase and phosphomannomutase activities. occur in low phosphate media (Southam et al. 1990). In In contrast, deletion of the gene encoding the putative M. mar- Hfx. volcanii, the transcription of the agl genes involved ipaludis acetyltransferase MMP0350 impaired flagellin N- in N-glycosylation occurs in a coordinated yet differential glycosylation (VanDyke et al. 2008), a process known to involve manner in the face of different growth paradigms, pointing acetylated sugars (Kelly et al. 2009). to this posttranslational modification as serving an adaptive Finally, in Pyrolobus fumarii, growing optimally at 106°C role (Yurist-Doutsch et al. 2008; Yurist-Doutsch et al. 2010). (Blöchl et al. 1997), novel UDP-sugars, including UDP-β- In addition to possibility affording advantages in the face GlcNAc-3-NAc and UDP-β-GlcNAc3NAc-(4-1)-β- of environmental challenges, archaeal N-glycosylation has GlcpNAc3NAc, have been reported (Gonçalves et al. 2008). been cited as providing structural support. This is best ex- Nothing, however, is known of their biosynthesis, whether emplified in Hbt. salinarum, where bacitracin treatment they are employed in N-glycosylation or, indeed, whether P. transformed rod-shaped cells into spheres (Mescher and Downloaded from fumarii even performs this posttranslational modification. Strominger 1976b). In cell wall-lacking T. acidophilum, the glycan moieties of the major membrane glycoprotein coating the cell surface have been suggested to either trap Roles of N-glycosylation in Archaea water molecules or encourage interaction between cell surface proteins. In either case, protein glycosylation would Whereas numerous N-glycosylated proteins have been de- http://glycob.oxfordjournals.org/ contribute to cell surface rigidity (Yang and Haug 1979b). scribed in a variety of Archaea living across a wide range of biological niches, the roles served by this posttranslational N-glycosylation of archaeal proteins has also been impli- modification remain, for the most part, unexplored. Indeed, cated in protein assembly and function. Defective Hfx. as aglB can be deleted from both Hfx. volcanii and M. voltae volcanii N-glycosylation resulted in an unstructured S-layer, (Abu-Qarn and Eichler 2006; Chaban, Voisin, et al. 2006), it comprised solely of the S-layer glycoprotein (Sumper et al. would seem that N-glycosylation is not essential for the surviv- 1990), while the absence of N-glycosylation compromised S- al of these species, at least under the conditions tested. layer stability (Abu-Qarn et al. 2007). Missing or partial M. voltae flagellin N-glycosylation led to a lack or reduction in Nonetheless, N-glycosylation may contribute to the ability of fl Archaea and their proteins to survive or adapt to the harsh en- agella numbers, concomitant with motility defects (Chaban, Voisin, et al. 2006). Likewise, bacitracin-mediated interference at Ben-Gurion University of the Negev, Aranne Library on November 18, 2013 vironments in which these organisms can thrive. fl In comparing the N-linked glycan profiles of two ha- with Methanococcus deltae agellin glycosylation resulted in a fl ff loarchaeal S-layer glycoproteins (Mengele and Sumper loss of agellation (Kalmoko et al. 1992). In mutant Hbt. sal- fl 1992), it was noted that the Hbt. salinarum protein not only inarum cells over-producing under-glycosylated agellins, fl experiences a higher degree of N-glycosylation than does increased levels of agella were detected in the growth medi- fl the same protein in Hfx. volcanii but that the glycans of um, implying that proper agellin glycosylation is necessary ff fl former are enriched in sulfated glucuronic acids, in contrast for e ective agellar incorporation into the plasma membrane to the neutral sugars found in the latter. Thus, relative to its (Wieland et al. 1985). In addition, glycosylation appears to fl Hfx. volcanii counterpart, the Hbt. salinarum S-layer glyco- play a role in stabilization against proteolysis, as re ected by protein presents a drastically increased surface charge the loss of Hfx. volcanii S-layer glycoprotein resistance to density, a property thought to contribute to the stability of added protease in mutants lacking many of the Agl proteins haloarchaeal proteins in the face of molar salt concentrations involved in assembling and attaching the pentasaccharide (Madern et al. 2000). Accordingly, the Hbt. salinarum S- N-linked to the protein (Yurist-Doutsch et al. 2008; Yurist- layer glycoprotein also contains 20% more acidic amino ac- Doutsch et al. 2010). Finally, N-glycosylation has been pro- id residues than does the Hfx. volcanii S-layer glycoprotein posed to modulate the interaction of binding proteins with (Lechner and Sumper 1987; Sumper et al. 1990). As a re- the cell membrane or envelope of S. acidocaldarius (Albers sult of these considerations, Hbt. salinarum is able to grow et al. 2004), while in Methanothermus sociabilis, the concept in higher salt concentrations than does Hfx. volcanii.InHfx. of N-glycosylation playing a role in cell aggregation has been volcanii, however, absent or defective N-glycosylation great- raised (Kärcher et al. 1993). ly reduced the ability of cells to grow at increasing salt concentrations (Abu-Qarn et al. 2007). The protection that Archaeal O-glycosylation an enhanced negative surface charge could also afford in the face of acidic conditions has been suggested as the rea- In addition to N-glycosylation, archaeal proteins can also expe- son for N-glycosylation of S. acidocaldarius cytochrome rience O-glycosylation. In the cases of both the Hbt. salinarum b558/566 (Hettmann et al. 1998; Zähringer et al. 2000). More- and the Hfx. volcanii S-layer glycoproteins, Thr-rich regions over, it was proposed that much of the protein surface is adjacent to the predicted membrane-spanning domain of the shielded from the ∼pH 2 environment in which these cells protein are modified with galactose–glucose disaccharides exist by the high degree of N-glycosylation it experiences (Mescher and Strominger 1976a; Sumper et al. 1990). Unfor- (Zähringer et al. 2000). It has also been postulated that tunately, virtually nothing is known of the archaeal O- the N-linked glycan of the M. fervidus S-layer glycoprotein glycosylation pathway at present. is involved in the stabilization of this surface protein at high temperatures (Kärcher et al. 1993). Available evidence also Concluding remarks points to Archaea as being able to modulate the N-glycosylation profile of target proteins. For instance, glycosylation Although initially described over 30 years, only recently of Methanospirillum hungatei flagellins was reported to only have major advances in delineating the archaeal pathway

1073 D Calo et al. of N-glycosylation pathway been made. What has been learned Bell SD, Jackson SP. 2001. Mechanism and regulation of transcription in ar- – thus far hints that continued examination of the archaeal ver- chaea. Curr Opin Microbiol. 4:208 213. ﬁ Ben-Dor S, Esterman N, Rubin E, Sharon N. 2004. Biases and complex pat- sion of this universal posttranslational modi cation will terns in the residues ﬂanking protein N-glycosylation sites. Glycobiology. provide insight into the process not obtained studying either 14:95–101. the parallel and better-characterized eukaryal or bacterial Blöchl E, Rachel R, Burggraf S, Hafenbradl D, Jannasch HW, Stetter KO. pathways. Moreover, as is so often the case when a given 1997. Pyrolobus fumarii, gen. and sp. nov., represents a novel group of archaea, extending the upper temperature limit for life to 113 degrees C. biological process is considered from the archaeal perspec- Extremophiles. 1:14–21. tive, analysis of N-glycosylation in a limited number of Brock TD, Brock KM, Belly RT, Weiss RL. 1972. Sulfolobus: A new genus of model systems has revealed facets of the process seemingly sulfur-oxidizing bacteria living at low pH and high temperature. Arch Micro- unique to members of this domain of life. On the other hand, biol. 84:54–68. Bult CJ, White O, Olsen GJ, Zhou L, Fleischmann RD, Sutton GG, Blake JA, particularly from the perspective of structural biology, the Downloaded from FitzGerald LM, Clayton RA, Gocayne JD, et al. 1996. Complete genome study of N-glycosylation in Archaea has served to elucidate sequence of the methanogenic archaeon, Methanaococcus jannaschii. Sci- aspects of the process common to all three domains of life. ence. 271:1058–1072. The on-going development of improved molecular tools for Burda P, Aebi M. 1999. The dolichol pathway of N-linked glycosylation. Bio- working with a variety of archaeal strains growing across a chim Biophys Acta. 1426:239–257. range of environmental conditions ensures that these unusual Cantarel BL, Coutinho PM, Rancurel C, Bernard T, Lombard V, Henrissat B.

2009. The Carbohydrate-Active EnZymes database (CAZy): An expert re- http://glycob.oxfordjournals.org/ microorganisms will continue to expand our understanding of source for Glycogenomics. Nucleic Acids Res. 37:D233–D238. N-glycosylation and other protein processing events. Chaban B, Logan SM, Kelly JF, Jarrell KF. 2009. AglC and AglK are involved in biosynthesis and attachment of diacetylated glucuronic acid to the N- glycan in Methanococcus voltae. J Bacteriol. 191:187–195. Funding Chaban B, Ng SY, Jarrell KF. 2006. Archaeal habitats–from the extreme to the ordinary. Can J Microbiol. 52:73–116. Research in the Eichler laboratory is supported by the Israel Chaban B, Voisin S, Kelly J, Logan SM, Jarrell KF. 2006. Identiﬁcation of Science Foundation (grant 30/07) and the US Army genes involved in the biosynthesis and attachment of Methanococcus voltae ﬃ N-linked glycans: Insight into N-linked glycosylation pathways in Archaea. Research O ce (grant W911NF-07-1-0260). L.K. is the Mol Microbiol. 61:259–268. recipient of a Negev-Zin Associates Scholarship. Dale H, Krebs MP. 1999. Membrane insertion kinetics of a protein domain in

vivo. The bacterioopsin N terminus inserts co-translationally. J Biol Chem. at Ben-Gurion University of the Negev, Aranne Library on November 18, 2013 274: 22693–22698. Conflict of interest statement DeLong EF. 1998. Everything in moderation: Archaea as ‘non-extremophiles’. Curr Opin Genet Dev. 8:649–654. None declared. Eichler J. 2001. Post-translational modification of the S-layer glycoprotein occurs following translocation across the plasma membrane of the haloarchaeon Haloferax volcanii. Eur J Biochem. 268:4366–4373. fi Abbreviations Eichler J, Adams MWW. 2005. Post-translational protein modi cation in Ar- chaea. Microbiol Mol Biol Rev. 69:393–425. ER, endoplasmic reticulum; GalNAc, N-acetylgalactosamine; Fine A, Irihimovitch V, Dahan I, Konrad Z, Eichler J. 2006. Cloning, expression, and purification of functional Sec11a and Sec11b, type I GlcNAc, N-acetylglucosamine; GT, glycosyltransferase; signal peptidases of the archaeon Haloferax volcanii. JBacteriol. OST, oligosaccharyltransferase; rRNA, ribosomal RNA; 188:1911–1919. S-layer, surface layer; UDP, uridine diphosphate; UTP, uri- Fox GE, Magrum LJ, Balch WE, Wolfe RS, Woese CR. 1977. Classification of dine triphosphate. methanogenic bacteria by 16S ribosomal RNA characterization. Proc Natl Acad Sci USA. 74:4537–4541. Francis CA, Beman JM, Kuypers MM. 2007. New processes and players in the References nitrogen cycle: The microbial ecology of anaerobic and archaeal ammonia oxidation. ISME J. 1:19–27. Abu-Qarn M, Eichler J. 2006. Protein N-glycosylation in Archaea: Defining Galperin MY. 2007. Using archaeal genomics to fight global warming and clos- Haloferax volcanii genes involved in S-layer glycoprotein glycosylation. tridia to fight cancer. Environ Microbiol. 9:279–286. Mol Microbiol. 61:511–525. Glover KJ, Weerapana E, Numao S, Imperiali B. 2005. Chemoenzymatic syn- Abu-Qarn M, Eichler J. 2007. An analysis of amino acid sequences surround- thesis of glycopeptides with PglB, a bacterial oligosaccharyl transferase ing archaeal glycoprotein sequons. Archaea. 2:73–81. from Campylobacter jejuni. Chem Biol. 12:1311–1315. Abu-Qarn M, Eichler J, Sharon N. 2008. Not just for Eukarya anymore: N- Gonçalves LG, Lamosa P, Huber R, Santos H. 2008. Di-myo-inositol phos- glycosylation in Bacteria and Archaea. Curr Opin Struct Biol. 18:544–550. phate and novel UDP-sugars accumulate in the extreme hyperthermophile Abu-Qarn M, Giordano A, Battaglia F, Trauner A, Hitchen P, Morris HR, Dell Pyrolobus fumarii. Extremophiles. 12:383–389. A, Eichler J. 2008. Identification of AglE, a second glycosyltransferase in- Gropp R, Gropp F, Betlach MC. 1992. Association of the halobacterial 7S volved in N-glycosylation of the Haloferax volcanii S-layer glycoprotein. J RNA to the polysome correlates with expression of the membrane protein Bacteriol. 190:3140–3146. bacterioopsin. Proc Natl Acad Sci USA. 89:1204–1208. Abu-Qarn M, Yurist-Doutsch S, Giordano A, Trauner A, Morris HR, Hitchen Hartmann E, Konig H. 1989. Uridine and dolichyl diphosphate activated P, Medalia O, Dell A, Eichler J. 2007. Haloferax volcanii AglB and AglD oligosaccharides are intermediates in the biosynthesis of the S-layer glyco- are involved in N-glycosylation of the S-layer glycoprotein and proper as- protein of Methanothermus fervidus. Arch Microbiol. 151:274–281. sembly of the surface layer. J Mol Biol. 374:1224–1236. Hettmann T, Schmidt CL, Anemüller S, Zähringer U, Moll H, Petersen A, Akutsu JI, Zhang Z, Tsujimura M, Sasaki M, Yohda M, Kawarabayasi Y. 2005. Schäfer G. 1998. Cytochrome b558/566 from the archaeon Sulfolobus acid- Characterization of a thermostable enzyme with phosphomannomutase/ ocaldarius. A novel highly glycosylated, membrane-bound b-type phosphoglucomutase activities from the hyperthermophilic archaeon Pyro- hemoprotein. J Biol Chem. 273:12032–12040. coccus horikoshii OT3. J Biochem. 138:159–166. Huber H, Hohn MJ, Rachel R, Fuchs T, Wimmer VC, Stetter KO. 2002. A new Albers SV, Koning SM, Konings WN, Driessen AJ. 2004. Insights into ABC phylum of Archaea represented by a nanosized hyperthermophilic symbiont. transport in archaea. J Bioenerg Biomembr. 36:5–15. Nature. 417:63–67. Amano T, Wakagi T, Oshima T. 1993. An ecto-enzyme from Sulfolobus acid- Igura M, Maita N, Kamishikiryo J, Yamada M, Obita T, Maenaka K, Kohda D. ocaldarius strain 7 which catalyzes hydrolysis of inorganic pyrophosphate, 2008. Structure-guided identification of a new catalytic motif of oligosac- ATP, and ADP: Purification and characterization. J Biochem. 114:329–333. charyltransferase. EMBO J. 27:234–243.

1074 Archaeal N-glycosylation

Imhoff JF, Kushner DJ, Kushwaha SC, Kates M. 1982. Polar lipids in photo- Mescher MF, Strominger JL. 1976b. Structural (shape-maintaining) role of the trophic bacteria of the Rhodospirillaceae and Chromatiaceae families. cell surface glycoprotein of Halobacterium salinarium. Proc Natl Acad Sci J Bacteriol. 150:1192–1201. USA. 73:2687–2691. Kalmokoff ML, Koval SF, Jarrell KF. 1992. Relatedness of the flagellins from Mescher MF, Strominger JL. 1978. Glycosylation of the surface glycoprotein methanogens. Arch Microbiol. 157:481–487. of Halobacterium salinarium via a cyclic pathway of lipid-linked intermedi- Kärcher U, Schröder H, Haslinger E, Allmaier G, Schreiner R, Wieland F, ates. FEBS Lett. 89:37–41. Haselbeck A, König H. 1993. Primary structure of the heterosaccharide Mescher MF, Hansen U, Strominger JL. 1976. Formation of lipid-linked sugar of the surface glycoprotein of Methanothermus fervidus. JBiolChem. compounds in Halobacterium salinarium. Presumed intermediates in glyco- 268:26821–26826. protein synthesis. J Biol Chem. 251:7289–7294. Kelly J, Logan SM, Jarrell KF, VanDyke DJ, Vinogradov E. 2009. A novel N- Meyer W, Schäfer G. 1992. Characterization and purification of a membrane- linked flagellar glycan from Methanococcus maripaludis. Carbohydr Res. bound archaebacterial pyrophosphatase from Sulfolobus acidocaldarius. 344:648–653. Eur J Biochem. 207:741–746. Klenk HP, Clayton RA, Tomb JF, White O, Nelson KE, Ketchum KA, Dodson Milewski S, Gabriel I, Olchowy J. 2006. Enzymes of UDP-GlcNAc biosynthe- RJ, Gwinn M, Hickey EK, Peterson JD, et al. 1997. The complete genomic sis in yeast. Yeast. 23:1–14. Downloaded from sequence of the hyperthermophilic, sulphate-reducing archaeon Archaeoglo- Mizanur RM, Zea CJ, Pohl NL. 2004. Unusually broad substrate tolerance of a bus fulgidus. Nature. 390:364–370. heat-stable archaeal sugar nucleotidyltransferase for the synthesis of sugar Konrad Z, Eichler J. 2002. Lipid modification of proteins in Archaea: Attach- nucleotides. J Am Chem Soc. 126:15993–15998. ment of a mevalonic acid-based lipid moiety to the surface-layer Mullakhanbhai MF, Larsen H. 1975. Halobacterium volcanii spec. nov., a glycoprotein of Haloferax volcanii follows protein translocation. Biochem Dead Sea halobacterium with a moderate salt requirement. Arch Microbiol. – – J. 366:959 964. 104:207 214. http://glycob.oxfordjournals.org/ Koonin EV, Wolf YI. 2008. Genomics of bacteria and archaea: The emerging Munoz MD, Hernandez LM, Basco R, Andaluz E, Larriba G. 1994. Glycosyl- dynamic view of the prokaryotic world. Nucleic Acids Res. 36:6688–6719. ation of yeast exoglucanase sequons in alg mutants deficient in the Kowarik M, Young NM, Numao S, Schulz BL, Hug I, Callewaert N, Mills DC, glucosylation steps of the lipid-linked oligosaccharide. Presence of gluco- Watson DC, Hernandez M, Kelly JF, et al. 2006. Definition of the bacterial triose unit in Dol-PP-GlcNAc2Man9Glc3 influences both glycosylation N-glycosylation site consensus sequence. EMBO J. 25:1957–1966. efficiency and selection of N-linked sites. Biochim Biophys Acta. Kuntz C, Sonnenbichler J, Sonnenbichler I, Sumper M, Zeitler R. 1997. Isola- 1201:361–366. tion and characterization of dolichol-linked oligosaccharides from Haloferax Namboori SC, Graham DE. 2008. Acetamido sugar biosynthesis in the eur- volcanii. Glycobiology. 7:897–904. yarchaea. J Bacteriol. 190:2987–2996. Lairson LL, Henrissat B, Davies GJ, Withers SG. 2008. Glycosyltransferases: Neuberger A. 1938. Carbohydrates in protein: The carbohydrate component of Structures, functions, and mechanisms. Annu Rev Biochem. 77:521–555. crystalline egg albumin. Biochem J. 32:1435–1451. Lechner J, Sumper M. 1987. The primary structure of a procaryotic glycopro- Paper W, Jahn U, Hohn MJ, Kronner M, Näther DJ, Burghardt T, Rachel R,

tein. Cloning and sequencing of the cell surface glycoprotein gene of Stetter KO, Huber H. 2007. Ignicoccus hospitalis sp. nov., the host of at Ben-Gurion University of the Negev, Aranne Library on November 18, 2013 halobacteria. J Biol Chem. 262:9724–9729. ‘Nanoarchaeum equitans’. Int J Syst Evol Microbiol. 57:803–808. Lechner J, Wieland F. 1989. Structure and biosynthesis of prokaryotic glyco- Petrescu AJ, Milac AL, Petrescu SM, Dwek RA, Wormald MR. 2004. Sta- proteins. Annu Rev Biochem. 58:173–194. tistical analysis of the protein environment of N-glycosylation sites: Lechner J, Wieland F, Sumper M. 1985a. Biosynthesis of sulfated saccharides Implications for occupancy, structure, and folding. Glycobiology. N-glycosidically linked to the protein via glucose. Purification and identifi- 14:103–114. cation of sulfated dolichyl monophosphoryl tetrasaccharides from Rashid N, Kanai T, Atomi H, Imanaka T. 2004. Among multiple phosphoman- halobacteria. J Biol Chem. 260:860–866. nomutase gene orthologues, only one gene encodes a protein with Lechner J, Wieland F, Sumper M. 1985b. Transient methylation of dolichyl phosphoglucomutase and phosphomannomutase activities in Thermococcus oligosaccharides is an obligatory step in halobacterial sulfated glycoprotein kodakaraensis. J Bacteriol. 186:6070–6076. biosynthesis. J Biol Chem. 260:8984–8989. Ray WK, Keith SM, DeSantis AM, Hunt JP, Larson TJ, Helm RF, Ken- Linton D, Dorrell N, Hitchen PG, Amber S, Karlyshev AV, Morris HR, Dell A, nelly PJ. 2005. A phosphohexomutase from the archaeon Sulfolobus Valvano MA, Aebi M, Wren BW. 2005. Functional analysis of the Campylo- solfataricus is covalently modified by phosphorylation on serine. J Bac- bacter jejuni N-linked protein glycosylation pathway. Mol Microbiol. teriol. 187:4270–4275. 55:1695–1703. Ring G, Eichler J. 2004. In the Archaea Haloferax volcanii, membrane protein Madern D, Ebel C, Zaccai G. 2000. Halophilic adaptation of enzymes. Extre- biogenesis and protein synthesis rates are affected by decreased ribosomal- mophiles. 4:91–98. binding to the translocon. J Biol Chem. 279:53160–53166. Magidovich H, Eichler J. 2009. Glycosyltransferases and oligosaccharyltrans- Rothschild LJ, Mancinelli RL. 2001. Life in extreme environments. Nature. ferases in Archaea: Putative components of the N-glycosylation pathway in 409:1092–1101. the third domain of life. FEMS Microbiol Lett. 300:122–130. Sandman K, Reeve JN. 2000. Structure and functional relationships of Magidovich H, Yurist-Doutsch S, Konrad Z, Ventura VV, Hitchen PG, Dell A, archaeal and eukaryal histones and nucleosomes. Arch Microbiol. Eichler J. 2010. AglP is a cytoplasmic S-adenosyl-L-methionine-dependent 173:165–169. methyltransferase that participates in the N-glycosylation pathway of Halo- Schreiner R, Schnabel E, Wieland F. 1994. Novel N-glycosylation in eukar- ferax volcanii. Mol Microbiol. 76:190–199. yotes: Laminin contains the linkage unit beta-glucosylasparagine. JCell Maita N, Nyirenda J, Igura M, Kamishikiryo J, Kohda D. 2010. Comparative Biol. 124:1071–1081. structural biology of eubacterial and archaeal oligosaccharyltransferases. Shams-Eldin H, Chaban B, Niehus S, Schwarz RT, Jarrell KF. 2008. Identi- J Biol Chem. 285:4941–4950. fication of the archaeal alg7 gene homolog (encoding N-acetylglucosamine- Manabe F, Itoh YH, Shoun H, Wakagi T. 2009. Membrane-bound acid pyro- 1-phosphate transferase) of the N-linked glycosylation system by phosphatase from Sulfolobus tokodaii, a thermoacidophilic archaeon: cross-domain complementation in Saccharomyces cerevisiae. J Bacteriol. Heterologous expression of the gene and characterization of the product. 190:2217–2220. Extremophiles. 13:859–865. Southam G, Kalmoko ML, Jarrell KF, Koval SF, Beveridge TJ. 1990. Iso- Mengele R, Sumper M. 1992. Drastic differences in glycosylation of related lation, characterization and cellular insertion of the flagella from two S-layer glycoproteins from moderate and extreme halophiles. J Biol Chem. strains of the archaebacterium Methanospirillum hungatei. J Bacteriol. 267:8182–8185. 172:3221–3228. Mengin-Lecreulx D, van Heijenoort J. 1994. Copurification of glucosamine-1- Spiro RG. 1973. Glycoproteins. Adv Protein Chem. 27:349–467. phosphate acetyltransferase and N-acetylglucosamine-1-phosphate uridyl- Sprott GD. 1992. Structures of archaebacterial membrane lipids. J Bioenerg transferase activities of Escherichia coli: Characterization of the glmU Biomembr. 24:555–566. gene product as a bifunctional enzyme catalyzing two subsequent steps in Stone KJ, Strominger JL. 1971. Mechanism of action of bacitracin: Complex- the pathway for UDP-N-acetylglucosamine synthesis. J Bacteriol. ation with metal ion and C 55-isoprenyl pyrophosphate. Proc Natl Acad Sci 176:5788–5795. USA. 68:3223–3227. Mescher MF, Strominger JL. 1976a. Purification and characterization of a pro- Sumper M, Berg E, Mengele R, Strobel I. 1990. Primary structure and gly- karyotic glucoprotein from the cell envelope of Halobacterium salinarium. cosylation of the S-layer protein of Haloferax volcanii. JBacteriol. J Biol Chem. 251:2005–2014. 172:7111–7118.

1075 D Calo et al.

Szymanski CM, Wren BW. 2005. Protein glycosylation in bacterial mucosal Yang LL, Haug A. 1979a. Purification and partial characterization of a procar- pathogens. Nat Rev Microbiol. 3:225–237. yotic glycoprotein from the plasma membrane of Thermoplasma Thauer RK, Kaster AK, Seedorf H, Buckel W, Hedderich R. 2008. Methano- acidophilum. Biochim Biophys Acta. 556:265–277. genic archaea: Ecologically relevant differences in energy conservation. Nat Yang LL, Haug A. 1979b. Structure of membrane lipids and physico-bio- Rev Microbiol. 6:579–591. chemical properties of the plasma membrane from Thermoplasma Turco SJ, Stetson B, Robbins PW. 1977. Comparative rates of transfer of lipid- acidophilum, adapted to growth at 37 degrees C. Biochim Biophys Acta. linked oligosaccharides to endogenous glycoprotein acceptors in vitro. Proc 573:308–320. Natl Acad Sci USA. 74:4411–4414. Young NM, Brisson JR, Kelly J, Watson DC, Tessier L, Lanthier PH, Jarrell VanDyke DJ, Wu J, Logan SM, Kelly JF, Mizuno S, Aizawa SI, Jarrell KF. HC, Cadotte N, St Michael F, Aberg E, et al. 2002. Structure of the N-linked 2009. Identification of genes involved in the assembly and attachment of glycan present on multiple glycoproteins in the Gram-negative bacterium, a novel flagellin N-linked tetrasaccharide important for motility in the ar- Campylobacter jejuni. J Biol Chem. 277:42530–42539. chaeon Methanococcus maripaludis. Mol Microbiol. 72:633–644. Yurist-Doutsch S, Abu-Qarn M, Battaglia F, Morris HR, Hitchen PG, Dell A, VanDyke DJ, Wu J, Ng SY, Kanbe M, Chaban B, Aizawa S, Jarrell KF. 2008. Eichler J. 2008. aglF, aglG and aglI, novel members of a gene cluster in- Identification of a putative acetyltransferase gene, MMP0350, which affects volved in the N-glycosylation of the Haloferax volcanii S-layer Downloaded from proper assembly of both flagella and pili in the archaeon Methanococcus glycoprotein. Mol Microbiol. 69:1234–1245. maripaludis. J Bacteriol. 190:5300–5307. Yurist-Doutsch S, Eichler J. 2009. Manual annotation, transcriptional analysis Voisin S, Houliston RS, Kelly J, Brisson JR, Watson D, Bardy SL, Jarrell KF, and protein expression studies reveal novel genes in the agl cluster respon- Logan SM. 2005. Identification and characterization of the unique N-linked sible for N-glycosylation in the halophilic archaeon Haloferax volcanii. J glycan common to the flagellins and S-layer glycoprotein of Methanococcus Bacteriol. 191:3068–3075. – voltae. J Biol Chem. 280:16586 16593. Yurist-Doutsch S, Magidovich H, Ventura VV, Hitchen PG, Dell A, Eichler J. http://glycob.oxfordjournals.org/ Wacker M, Linton D, Hitchen PG, Nita-Lazar M, Haslam SM, North SJ, Pa- 2010. N-glycosylation in Archaea: On the coordinated actions of Haloferax nico M, Morris HR, Dell A, Wren BW, et al. 2002. N-linked glycosylation volcanii AglF and AglM. Mol Microbiol. 75:1047–1058. in Campylobacter jejuni and its functional transfer into E. coli. Science. Zähringer U, Moll H, Hettmann T, Knirel YA, Schäfer G. 2000. Cytochrome 298:1790–1793. b558/566 from the archaeon Sulfolobus acidocaldarius has a unique Asn- Waters E, Hohn MJ, Ahel I, Graham DE, Adams MD, Barnstead M, Beeson linked highly branched hexasaccharide chain containing 6-sulfoquinovose. KY, Bibbs L, Bolanos R, Keller M, et al. 2003. The genome of Nanoarch- Eur J Biochem. 267:4144–4149. aeum equitans: Insights into early archaeal evolution and derived parasitism. Zeitler R, Hochmuth E, Deutzmann R, Sumper M. 1998. Exchange of Ser-4 for Proc Natl Acad Sci USA. 100:12984–12988. Val, Leu or Asn in the sequon Asn-Ala-Ser does not prevent N-glycosylation Weerapana E, Imperiali B. 2006. Asparagine-linked protein glycosylation: of the cell surface glycoprotein from Halobacterium halobium. Glycobiol- From eukaryotic to prokaryotic systems. Glycobiology. 16:91R–101R. ogy. 8:1157–1164. Wieland F, Dompert W, Bernhardt G, Sumper M. 1980. Halobacterial glycopro- Zhu BC, Drake RR, Schweingruber H, Laine RA. 1995. Inhibition of glyco-

tein saccharides contain covalently linked sulphate. FEBS Lett. 120:110–114. sylation by amphomycin and sugar nucleotide analogs PP36 and PP55 at Ben-Gurion University of the Negev, Aranne Library on November 18, 2013 Wieland F, Heitzer R, Schaefer W. 1983. Asparaginylglucose: Novel type of indicates that Haloferax volcanii beta-glucosylates both glycoproteins and carbohydrate linkage. Proc Natl Acad Sci USA. 80:5470–5474. glycolipids through lipid-linked sugar intermediates: Evidence for three nov- Wieland F, Paul G, Sumper M. 1985. Halobacterial ﬂagellins are sulfated gly- el glycoproteins and a novel sulfated dihexosyl-archaeol glycolipid. Arch coproteins. J Biol Chem. 260:15180–15185. Biochem Biophys. 319: 355–364. Woese CR, Fox GE. 1977. Phylogenetic structure of the prokaryotic domain: Zhu BC, Laine RA. 1996. Dolichyl-phosphomannose synthase from the archae The primary kingdoms. Proc Natl Acad Sci USA. 74:5088–5090. Thermoplasma acidophilum. Glycobiology. 6:811–816. Yan Q, Lennarz WJ. 2002. Studies on the function of oligosaccharyl transferase Zuﬀerey R, Knauer R, Burda P, Stagljar I, te Heesen S, Lehle L, Aebi M. 1995. subunits. Stt3p is directly involved in the glycosylation process. JBiol STT3, a highly conserved protein required for yeast oligosaccharyl transfer- Chem. 277:47692–47700. ase activity in vivo. EMBO J. 14: 4949–4960.

1076

Appendix C

"Add salt, add sugar: N-glycosylation in Haloferax volcanii"

Lina Kaminski, Shai Naparstek, Lina Kandiba, Chen Cohen-Rosenzweig, Adi Arbiv, Zvia Konrad, and Jerry Eichler Biochem. Soc. Trans. 2013c. 41:432-435

432 Biochemical Society Transactions (2013) Volume 41, part 1

Add salt, add sugar: N-glycosylation in Haloferax volcanii

Lina Kaminski, Shai Naparstek, Lina Kandiba, Chen Cohen-Rosenzweig, Adi Arbiv, Zvia Konrad and Jerry Eichler1 Department of Life Sciences, Ben Gurion University of the Negev, Beersheva 84105, Israel

Abstract Although performed by members of all three domains of life, the archaeal version of N-glycosylation remains the least understood. Studies on Haloferax volcanii have, however, begun to correct this situation. A combination of bioinformatics, molecular biology, biochemical and mass spectrometry approaches have served to delineate the Agl pathway responsible for N-glycosylation of the S-layer glycoprotein, a reporter of this post-translational modification in Hfx. volcanii. More recently, differential N-glycosylation of the S-layer glycoprotein as a function of environmental salinity was demonstrated, showing that this post-translational modification serves an adaptive role in Hfx. volcanii. Furthermore, manipulation of the Agl pathway, together with the capability of Hfx. volcanii to N-glycosylate nonnative proteins, forms the basis for establishing this species as a glyco-engineering platform. In the present review, these and other recent findings are addressed.

Introduction Although beyond the scope of the present review, it should Across evolution, analysis of the proteome reveals additional be stressed that our present understanding of archaeal N- levels of complexity not predicted at the genome level. glycosylation is not solely based on results obtained using Post-translational modifications are a major source of Hfx. volcanii as a model system. Indeed, recent studies of this proteomic diversity. Insight into how such protein- the methanogens Methanococcus voltae and Methanococcus processing events transpire in Archaea has come from studies maripaludis, and the thermophiles Sulfolobus acidocaldarius, on Haloferax volcanii, where numerous and varied examples Pyrococcus furiosus and Archaeoglobus fulgidus have also of post-translational modifications have been reported [1]. provided considerable insight into the process of N- N-glycosylation, the covalent linkage of glycan moieties to glycosylation in Archaea [8–13]. select asparagine residues of a target protein, was among the first post-translational modifications to be described in Hfx. volcanii. The Hfx. volcanii Agl pathway In Hfx. volcanii, the surface (S)-layer glycoprotein, Over the last few years, substantial progress in deciphering comprising the sole component of the S-layer surrounding the the pathway of Hfx. volcanii N-glycosylation has been cell, contains seven putative N-glycosylation sites, namely made stemming from the identification of a series of agl the Asn-Xaa-Ser/Thr sequence motif, where Xaa is any (archaeal glycosylation) genes encoding proteins involved in residue but proline [2]. Early studies had reported modi- the assembly and attachment of a pentasaccharide comprising fication of the S-layer glycoprotein by linear strings of a hexose, two hexuronic acids, the methyl ester of a hexuronic glucose residues, as well as by a second glycan containing acid and a mannose to select asparagine residues of the S- glucose, galactose and idose subunits [2,3]. The recent use layer glycoprotein. The agl genes were originally identified of more sophisticated MS tools has, however, served to on the basis of the homology of their protein products revise the composition of the N-linked glycans decorating with components of the better-defined eukaryal and bacterial the S-layer glycoprotein [4–6]. At the same time, studies N-glycosylation pathways [14]. Additional agl genes were combining gene deletion with lipid- and protein-linked later identified on the basis of their genomic proximities glycan analysis have served to delineate a pathway of N- to the first set of agl genes [15]. Indeed, all but one of glycosylation in Hfx. volcanii [7]. In the present review, the agl genes (i.e. aglD) are clustered into a single gene recent discoveries regarding Hfx. volcanii N-glycosylation island. Moreover, as the major protein species in Hfx. are discussed. In particular, the adaptive role of Hfx. volcanii volcanii, the S-layer glycoprotein represents a convenient N-glycosylation to changes in the environment and the reporter for studying the roles of the agl gene products in manipulation of the Hfx. volcanii N-glycosylation pathway N-glycosylation. for glyco-engineering purposes are addressed. Acting at the cytoplasmic face of the plasma membrane, AglJ, AglG, AglI and AglE are glycosyltransferases that Key words: Archaea, Haloferax volcanii, N-glycosylation, post-translational modiﬁcation, sequentially add the first four pentasaccharide residues on proteomic diversity. to a common DolP (dolichol phosphate) carrier, while Abbreviations used: agl, archaeal glycosylation; DolP, dolichol phosphate; ER, endoplasmic reticulum; S-layer, surface layer. AglD adds the final pentasaccharide residue, mannose, 1 To whom correspondence should be addressed (email [email protected]). to a distinct DolP [4–6,16–19]. The use of DolP as the

C C Biochemical Society Transactions www.biochemsoctrans.org The Authors Journal compilation 2013 Biochemical Society Biochem. Soc. Trans. (2013) 41, 432–435; doi:10.1042/BST20120142 Molecular Biology of Archaea 3 433

Figure 1 The current model of the Agl pathway responsible for N-glycosylation of the Hfx. volcanii S-layer glycoprotein at Asn13 and Asn83 See the text for details.

lipid carrier during N-linked glycan assembly is also the system capable of generating designer glycoproteins tailored case in eukaryal N-glycosylation [20]; bacterial N-linked for enhanced activity, stability or longevity in the face glycans are instead assembled on a different polyprenoid, of extreme conditions are warranted. Specifically, with a undecaprenol pyrophosphate [21,22]. In Hfx. volcanii,N- relatively well-defined N-glycosylation pathway and the glycosylation-related roles have also been assigned to AglF, availability of appropriate tools for genetic manipulation, a glucose-1-phosphate uridyltransferase [19], AglM, a UDP- Hfx. volcanii is a promising candidate upon which to glucose dehydrogenase [19], and AglP, a methyltransferase establish a glyco-engineering platform. However, before [5]. Indeed, AglF and AglM were shown to act in a such efforts can proceed, two criteria must be met. First, sequential and co-ordinated manner in vitro, transforming it is necessary to develop a series of Hfx. volcanii strains glucose 1-phosphate into UDP-glucuronic acid [19]. In capable of performing differential N-glycosylation. Secondly, a reaction requiring the archaeal oligosaccharyltransferase, it is necessary to show that Hfx. volcanii is capable of N- AglB [8,9,14], the lipid-linked tetrasaccharide and its glycosylating non-native proteins. Of late, progress has been precursors are delivered to select asparagine residues of made on both of these fronts. By replacing components of the S-layer glycoprotein. The final mannose residue is the Agl pathway with genes encoding homologous proteins subsequently transferred from its DolP carrier to the protein- from Halobacterium salinarum, Haloquadratum walsbyi or bound tetrasaccharide [6]. Current understanding of the Hfx. Haloarcula marismortui, Hfx. volcanii strains capable of volcanii Agl pathway is depicted in Figure 1. decorating the S-layer glycoprotein with N-linked glycans distinct from the pentasaccharide normally decorating this protein were created [25,26]. At the same time, it was shown that VP4 (viral protein 4), the major structural Hfx. volcanii as a glyco-engineering protein of HRPV-1 (Halorubrum pleomorphic virus 1), is platform N-glycosylated at the same sites when expressed in either Although N-glycosylation has only been addressed in a the native host, Halorubrum sp. strain PV6, or in Hfx. limited number of species, genomic analysis predicts this volcanii. The composition of the N-linked glycan in each case post-translational modification to be common in Archaea was, however, species-specific, with Hfx. volcanii adding the [23]. At the same time, the glycans N-linked to archaeal same pentasaccharide as N-linked to the S-layer glycoprotein glycoproteins present a diversity of sugar subunits not seen [27]. These proof-of-concept results justify the continuation in either Eukarya or Bacteria [24]. With these points in of efforts to develop Hfx. volcanii into a versatile glyco- mind, attempts at developing an archaeal glyco-engineering engineering platform.

C The Authors Journal compilation C 2013 Biochemical Society 434 Biochemical Society Transactions (2013) Volume 41, part 1

Different haloarchaea, different glycoprotein was compared in cells grown in 3.4 or 1.75 M N-glycosylation strategies NaCl-containing medium [31]. At the higher level of salinity, 13 83 The insight into archaeal N-glycosylation gained by studying S-layer glycoprotein Asn and Asn were shown to be Hfx. volcanii has served to elucidate aspects of this post- modified by the pentasaccharide described above, whereas translational modification in other Haloarchaea. Recent DolP was shown to be modified by the tetrasaccharide efforts revealed that the S-layer glycoprotein of both Hfx. comprising the first four pentasaccharide residues, again as volcanii and Har. marismortui are decorated with the same N- discussed above. In contrast, cells grown at the lower level of linked pentasaccharide [28]. Nonetheless, differences in the salinity were shown to contain DolP modified by a distinct N-glycosylation pathways of these two Haloarchaea exist. tetrasaccharide comprising a sulfated hexose, two hexoses Whereas the N-linked pentasaccharide decorating the Hfx. and a rhamnose. No such glycan was detected linked to volcanii S-layer glycoprotein is derived from a tetrasaccharide DolP in cells grown at the higher level of salinity. The same 498 sequentially assembled on a common DolP carrier and a final tetrasaccharide-modified S-layer glycoprotein Asn in cells mannose residue derived from a distinct DolP carrier [6], grown in low salt, whereas no glycan decorated this residue in 13 the same pentasaccharide N-linked to the Har. marismortui cells grown in the high salt medium. At the same time, Asn 83 S-layer glycoprotein is fully assembled on a single DolP, and Asn were modified by substantially less pentasaccharide from where it is transferred to the protein [28]. As such, under the low salt conditions. Hence, in response to the haloarchaeal N-glycosylation relies on pathways reminiscent degree of environmental salinity, Hfx. volcanii modulates not of either the parallel eukaryal or bacterial processes. only the composition of the N-linked glycans decorating the The Hfx. volcanii N-glycosylation pathway, involving S-layer glycoprotein, but also which residues undergo this multiple glycan-charged DolP carriers, recalls its counterpart post-translational modification. in higher Eukarya. Here, the first seven subunits of the 14-meric oligosaccharide assembled in the ER (endoplasmic What next? reticulum) are sequentially added to a common dolichol Although considerable advances in our understanding of pyrophosphate carrier; the second set of seven sugar subunits archaeal N-glycosylation have been realized using Hfx. are derived from single mannose- or glucose-charged DolP volcanii as a model system, numerous questions remain un- [20,29]. The finding that the final mannose of the N-linked answered. One of the more pressing open questions concerns pentasaccharide decorating the Hfx. volcanii S-layer gly- the identity of the flippase(s) responsible for translocating coprotein is added to the tetrasaccharide already attached to glycan-charged DolP across the plasma membrane during N- the protein [28] shows further the eukaryal-like nature of N- glycosylation. Indeed, little is known of lipid-linked oligosac- glycosylation in this organism. In Eukarya, N-glycosylation charide flippases in any glycosylation system. The agents re- begins with the delivery of a lipid-linked oligosaccharide to sponsible for several other predicted steps of the Hfx. volcanii the target protein in the ER. Modification continues in the Agl pathway also have yet to be reported, such as the enzyme Golgi, where additional sugar subunits are attached to responsible for delivering mannose from its DolP carrier the oligosaccharide already N-linked to the target protein. to the S-layer glycoprotein-bound tetrasaccharide. Further- By contrast, by relying on a single glycan-charged DolP more, little is known of DolP biosynthesis in Hfx. volcanii. carrier, N-glycosylation in Har. marismortui is similar to the The fact that the Hfx. volcanii S-layer glycoprotein parallel bacterial process, as exemplified by Campylobacter undergoes differential N-glycosylation as a function of jejuni. Here, a heptasaccharide is assembled by the sequential environmental salinity raises a distinct set of questions. addition of seven soluble nucleotide-activated sugars on to What advantage does this differential N-glycosylation offer a common undecaprenol phosphate carrier. The complete the cell? What enzymes contribute to the biogenesis of the heptasaccharide is then delivered to the target protein and ‘low-salt’ tetrasaccharide? Is AglB responsible for delivering does not undergo further processing [21,22,30]. the ‘low-salt’ tetrasaccharide from its DolP carrier to S- layer glycoprotein Asn498? If this is indeed the case, then N-glycosylation as an adaptive response Hfx. volcanii AglB would have to be more versatile that to environmental salinity any known oligosaccharyltransferase. Finally, one can ask whether additional variations in S-layer N-glycosylation Although perturbation of N-glycosylation compromises occur in response to other environmental conditions. the ability of Hfx. volcanii to grow in high salt, S-layer Clearly, further study of Hfx. volcanii N-glycosylation stability and architecture, and S-layer resistance to added will continue to reveal new twists on this universal post- protease [4,17–19], cells lacking AglB, and hence unable translational modification. The future indeed looks sweet. to perform N-glycosylation, are viable [4]. Thus, although not essential for Hfx. volcanii survival, N-glycosylation is, nonetheless, advantageous to Hfx. volcanii in certain Funding scenarios. As such, one can propose that Hfx. volcanii modifies aspects of N-glycosylation in response to changing J.E. was supported by the Israel Science Foundation (8/11) and the growth conditions. Recently, this hypothesis was tested and US Army Research Ofﬁce [grant number W911NF-11-1-520]. L.K. and shown to be true when N-glycosylation of the S-layer S.N. are recipients of a Negev-Zin Associates Scholarship.

C The Authors Journal compilation C 2013 Biochemical Society Molecular Biology of Archaea 3 435

References 16 Abu-Qarn, M., Giordano, A., Battaglia, F., Trauner, A., Morris, H.R., 1 Eichler, J. and Adams, M.W.W. (2005) Post-translational protein Hitchen, P., Dell, A. and Eichler, J. (2008) Identification of AglE, a second modification in Archaea. Microbiol. Mol. Biol. Rev. 69, 393–425 glycosyltransferase involved in N-glycosylation of the Haloferax volcanii 2 Sumper, M., Berg, E., Mengele, R. and Strobel, I. (1990) Primary structure S-layer glycoprotein. J. Bacteriol. 190, 3140–3146 and glycosylation of the S-layer protein of Haloferax volcanii.J.Bacteriol. 17 Yurist-Doutsch, S., Abu-Qarn, M., Battaglia, F., Morris, H.R., Hitchen, P.G., 172, 7111–7118 Dell, A. and Eichler, J. (2008) aglF, aglG and aglI, novel members of a 3 Mengele, R. and Sumper, M. (1992) Drastic differences in glycosylation gene cluster involved in the N-glycosylation of the Haloferax volcanii of related S-layer glycoproteins from moderate and extreme halophiles. S-layer glycoprotein. Mol. Microbiol. 69, 1234–1245 J. Biol. Chem. 267, 8182–8185 18 Kaminski, L., Abu-Qarn, M., Guan, Z., Naparstek, S., Ventura, V.V., Raetz, 4 Abu-Qarn, M., Yurist-Doutsch, S., Giordano, A., Trauner, A., Morris, H.R., C.R.H., Hitchen, P.G., Dell, A. and Eichler, J. (2010) AglJ adds the first Hitchen, P., Medalia, O., Dell, A. and Eichler, J. (2007) Haloferax sugar of the N-linked pentasaccharide decorating the Haloferax volcanii volcanii AglB and AglD are involved in N-glycosylation of the S-layer S-layer glycoprotein. J. Bacteriol. 192, 5572–5579 glycoprotein and proper assembly of the surface layer. J. Mol. Biol. 374, 19 Yurist-Doutsch, S., Magidovich, H., Ventura, V.V., Hitchen, P.G., Dell, A. 1224–1236 and Eichler, J. (2010) N-glycosylation in Archaea: on the coordinated 5 Magidovich, H., Yurist-Doutsch, S., Konrad, Z., Ventura, V.V., Dell, A., actions of Haloferax volcanii AglF and AglM. Mol. Microbiol. 75, Hitchen, P.G. and Eichler, J. (2010) AglP is a S-adenosyl-l-methionine- 1047–1058 dependent methyltransferase that participates in the N-glycosylation 20 Burda, P. and Aebi, M. (1999) The dolichol pathway of N-linked pathway of Haloferax volcanii. Mol. Microbiol. 76, 190–199 glycosylation. Biochim. Biophys. Acta 1426, 239–257 6 Guan, Z., Naparstek, S., Kaminski, L., Konrad, Z. and Eichler, J. (2010) 21 Szymanski, C.M. and Wren, B.W. (2005) Protein glycosylation in bacterial Distinct glycan-charged phosphodolichol carriers are required for the mucosal pathogens. Nat. Rev. Microbiol. 3, 225–237 assembly of the pentasaccharide N-linked to the Haloferax volcanii 22 Hartley, M.D. and Imperiali, B. (2011) At the membrane frontier: a S-layer glycoprotein. Mol. Microbiol. 78, 1294–1303 prospectus on the remarkable evolutionary conservation of polyprenols 7 Calo, D., Kaminski, L. and Eichler, J. (2010) Protein glycosylation in and polyprenyl-phosphates. Arch. Biochem. Biophys. 517, 83–97 Archaea: sweet and extreme. Glycobiology 20, 1065–1079 23 Magidovich, H. and Eichler, J. (2009) Glycosyltransferases and 8 Chaban, B., Voisin, S., Kelly, J., Logan, S.M. and Jarrell, K.F. (2006) oligosaccharyltransferases in Archaea: putative components of the Identification of genes involved in the biosynthesis and attachment of N-glycosylation pathway in the third domain of life. FEMS Microbiol. Lett. Methanococcus voltae N-linked glycans: insight into N-linked 300, 122–130 glycosylation pathways in Archaea. Mol. Microbiol. 61, 259–268 24 Schwarz, F. and Aebi, M. (2011) Mechanisms and principles of N-linked 9 Igura, M., Maita, N., Kamishikiryo, J., Yamada, M., Obita, T., Maenaka, K. protein glycosylation. Curr. Opin. Struct. Biol. 21, 576–582 and Kohda, D. (2008) Structure-guided identification of a new catalytic 25 Calo, D., Eilam, Y., Lichtenstein, R.G. and Eichler, J. (2010) Towards motif of oligosaccharyltransferase. EMBO J. 27, 234–243 glyco-engineering in Archaea: replacing Haloferax volcanii AglD with 10 VanDyke, D.J., Wu, J., Logan, S.M., Kelly, J.F., Mizuno, S., Aizawa, S. and homologous glycosyltransferases from other halophilic archaea. Jarrell, K.F. (2009) Identification of genes involved in the assembly and Appl. Environ. Microbiol. 76, 5684–5692 attachment of a novel flagellin N-linked tetrasaccharide important for 26 Calo, D., Guan, Z. and Eichler, J. (2011) Glyco-engineering in Archaea: motility in the archaeon Methanococcus maripaludis. Mol. Microbiol. 72, differential N-glycosylation of the S-layer glycoprotein in a transformed 633–644 Haloferax volcanii strain. Microbiol. Biotechnol. 4, 461–470 11 Meyer, B.H., Zolghadr, B., Peyfoon, E., Pabst, M., Panico, M., Morris, H.R., 27 Kandiba, L., Aitio, O., Helin, J., Guan, Z., Permi, P., Bamford, D., Eichler, J. Haslam, S.M., Messner, P., Schaffer, ¨ C., Dell, A. and Albers, S.V. and Roine, E. (2012) Diversity in prokaryotic glycosylation: an (2011) Sulfoquinovose synthase: an important enzyme in the archaeal-derived N-linked glycan contains legionaminic acid. Mol. N-glycosylation pathway of Sulfolobus acidocaldarius. Mol. Microbiol. Microbiol. 84, 578–593 82, 1150–1163 28 Calo, D., Guan, Z., Naparstek, S. and Eichler, J. (2011) Different routes to 12 Jones, G.M., Wu, J., Ding, Y., Uchida, K., Aizawa, S., Robotham, A., Logan, the same ending: comparing the N-glycosylation processes of Haloferax S.M., Kelly, J. and Jarrell, K.F. (2012) Identification of genes involved in volcanii and Haloarcula marismortui, two halophilic archaea from the the acetamidino group modification of the flagellin N-linked glycan of Dead Sea. Mol. Microbiol. 81, 1166–1177 Methanococcus maripaludis. J. Bacteriol. 194, 2693–2702 29 Helenius, A. and Aebi, M. (2004) Roles of N-linked glycans in the 13 Matsumoto, S., Igura, M., Nyirenda, J., Matsumoto, M., Yuzawa, S., Noda, endoplasmic reticulum. Annu. Rev. Biochem. 73, 1019–1049 N.N., Inagaki, F. and Kohda, D. (2012) Crystal structure of the C-terminal 30 Abu-Qarn, M., Eichler, J. and Sharon, N. (2008) Not just for Eukarya globular domain of oligosaccharyltransferase from Archaeoglobus anymore: N-glycosylation in Bacteria and Archaea. Curr. Opin. Struct. fulgidus at 1.75 Å resolution. Biochemistry 51, 4157–4166 Biol. 18, 544–550 14 Abu-Qarn, M. and Eichler, J. (2006) Protein N-glycosylation in Archaea: 31 Guan, Z., Naparstek, S., Calo, D. and Eichler, J. (2012) Protein defining Haloferax volcanii genes involved in S-layer glycoprotein glycosylation as an adaptive response in Archaea: growth at different glycosylation. Mol. Microbiol. 61, 511–525 salt concentrations leads to alterations in Haloferax volcanii S-layer 15 Yurist-Doutsch, S. and Eichler, J. (2009) Manual annotation, glycoprotein N-glycosylation. Environ. Microbiol. 14, 743–753 transcriptional analysis and protein expression studies reveal novel genes in the agl cluster responsible for N-glycosylation in the halophilic Received 7 June 2012 archaeon Haloferax volcanii. J. Bacteriol. 191, 3068–3075 doi:10.1042/BST20120142

C The Authors Journal compilation C 2013 Biochemical Society

5. Contributions to this thesis

All of the experiments described in this thesis were performed by me, except for the deletion of the HVO_1517, aglR and HVO_2048 genes, which was performed by

Dr. Mehtap Abu-Qarn and Dr. Sophie Yurist-Doutsch. Furthermore, the mass spectrometry experiments were performed in collaboration with Dr. Anne Dell in my first paper (Kaminski et al., 2010) and with Dr. Ziqang Guan on my second, third and fourth papers (Kaminski et al., 2010; Kaminski et al., 2012; Kaminski et al., 2013a).

6. Papers resulting from this thesis

1. Kaminski, L., Neparstek, S., Abu-Qarn, M., Eilam, Y., Guan, Z., Raetz, C.R.,

Lichtenstein, R. and Eichler, J. (2010). AglJ, a novel component of the Haloferax

volcanii N-glycosylation pathway. J. Bacteriol. 192:5572-5579

2. Kaminski, L. and Eichler, J. (2010). Identification of residues important for the

activity of Haloferax volcanii AglD, a component of the archaeal N-glycosylation

pathway. Archaea. 6;2010:315108

3. Kaminski, L., Guan, Z., Abu-Qarn, M., Konrad, Z. and Eichler, J. (2012). AglR

is required for addition of the final mannose residue of the N-linked glycan

decorating the Haloferax volcanii S-layer glycoprotein. Biochim. Biophys. Acta.

1820:1664-1670

4. Kaminski, L., Guan, Z., Yurist-Doutsch, S. and Eichler, J. (2013a) Two distinct

N-glycosylation pathways together process the S-layer glycoprotein in the

halophilic archaea, Haloferax volcanii. mBio. 5;4. pii: e00716-13

5. Kaminski, L., Lurie-Weinberger, MN., Allers, T., Gophna, U and Eichler, J.

(2013b) Phylogenetic- and genome-derived insight into the evolution of N-

glycosylation in Archaea. Mol. Phylogenet. Evol. 68:327-339.

6. Guan, Z., Naparstek, S., Kaminski, L., Konrad, Z. and Eichler, J. (2010).

Distinct glycan-charged phosphodolichol carriers are required for the assembly of

the pentasaccharide N-linked to the Haloferax volcanii S-layer glycoprotein. Mol.

Microbiol. 78:1294-1303

7. Calo, D. Kaminski, L. and Eichler, J. (2010). Protein glycosylation in Archaea:

Sweet and Extreme. Glycobiology. 20:1065-1076

8. Kaminski, L., Naparstek, S., Kandiba, L., Cohen-Rosenzweig, C., Arbiv, A.,

Konrad, Z. and Eichler, J. (2013c) Add salt, add sugar: N-glycosylation in

Haloferax volcanii. Biochem. Soc. Trans. 41:432-435.

7. References

Abu-Qarn, M., and Eichler, J. (2006) Protein N-glycosylation in Archaea: Defining Haloferax volcanii genes involved in S-layer glycoprotein glycosylation. Mol Microbiol 61: 511–25.

Abu-Qarn, M., Yurist-Doutsch, S., Giordano, A., Trauner, A., Morris, H.R., Hitchen, P.G., et al. (2007) Haloferax volcanii AglB and AglD are involved in N-glycosylation of the S-layer glycoprotein and proper assembly of the surface layer. J Mol Biol 374: 1224–1236.

Abu-Qarn, M., Giordano, A., Battaglia, F., Trauner, A., Hitchen, P.G., Morris, H.R., et al. (2008) Identification of AglE, a second glycosyltransferase involved in N- glycosylation of the Haloferax volcanii S-layer glycoprotein. J Bacteriol 190: 3140– 3146.

Alaimo, C., Catrein, I., Morf, L., Marolda, C.L., Callewaert, N., Valvano, M.A., et al. (2006) Two distinct but interchangeable mechanisms for flipping of lipid-linked oligosaccharides. Eur Mol Biol Organ J 25: 967–976.

Allers, T., Ngo, H., Mevarech, M., and Lloyd, R.G. (2004) Development of additional selectable markers for the halophilic archaeon Haloferax volcanii based on the leuB and trpA genes. Appl Environ Microbiol 70: 943–953.

Anand, M., Rush, J.S., Ray, S., Doucey, M.A., Weik, J., Ware, F.E., et al. (2001) Requirement of the Lec35 gene for all known classes of monosaccharide-P-dolichol- dependent glycosyltransferase reactions in mammals. Mol Biol Cell 12: 487–501.

Apweiler, R., Hermjakob, H., and Sharon, N. (1999) On the frequency of protein glycosylation, as deduced from analysis of the SWISS-PROT database. Biochim Biophys Acta 1473: 4–8.

Benning, C. (1998) Biosynthesis and function of the sulfolipid sulfoquinovosyl diacylglycerol. Annu Rev Plant Physiol Plant Mol Biol 49: 53–75.

Burda, P., and Aebi, M. (1999) The dolichol pathway of N-linked glycosylation. Biochim Biophys Acta 1426: 239–257.

Chaban, B., Logan, S.M., Kelly, J.F., and Jarrell, K.F. (2009) AglC and AglK are involved in biosynthesis and attachment of diacetylated glucuronic acid to the N- glycan in Methanococcus voltae. J Bacteriol 191: 187–195.

Chaban, B., Voisin, S., Kelly, J., Logan, S.M., and Jarrell, K.F. (2006) Identification of genes involved in the biosynthesis and attachment of Methanococcus voltae N- linked glycans: insight into N-linked glycosylation pathways in Archaea. Mol Microbiol 61: 259–268.

Cohen-Rosenzweig, C., Yurist-Doutsch, S., and Eichler, J. (2012) AglS, a novel component of the Haloferax volcanii N-glycosylation pathway, is a dolichol phosphate-mannose mannosyltransferase. J Bacteriol 194: 6909–6916.

Ding, Y., Jones, G.M., Uchida, K., Aizawa, S.I., Robotham, A., Logan, S.M., et al. (2013) Identification of genes involved in the biosynthesis of the third and fourth sugars of the Methanococcus maripaludis archaellin N-linked tetrasaccharide. J Bacteriol 195: 4094–4104.

Eichler, J. (2013) Extreme sweetness: protein glycosylation in archaea. Nat Rev Microbiol 11: 151–156.

Eichler, J., and Adams, M.W.W. (2005) Posttranslational protein modification in Archaea. Microbiol Mol Biol Rev 69: 393-425.

Feldman, M.F., Wacker, M., Hernandez, M., Hitchen, P.G., Marolda, C.L., Kowarik, M., et al. (2005) Engineering N-linked protein glycosylation with diverse O antigen lipopolysaccharide structures in Escherichia coli. Proc Natl Acad Sci U S A 102: 3016–3021.

Fox, G.E., Magrum, L.J., Balch, W.E., Wolfe, R.S., and Woese, C.R. (1977) Classification of methanogenic bacteria by 16S ribosomal RNA characterization. Proc Natl Acad Sci U S A 74: 4537–4541.

Frank, C.G., Sanyal, S., Rush, J.S., Waechter, C.J., and Menon, A.K. (2008) Does Rft1 flip an N-glycan lipid precursor? Nature 454: E3–E4; discussion E4–E5.

Garinot-Schneider, C., Lellouch, A.C., and Geremia, R.A. (2000) Identification of essential amino acid residues in the Sinorhizobium meliloti glucosyltransferase ExoM. J Biol Chem 275: 31407–31413.

Gavel, Y., and von Heijne, G. (1990) Sequence differences between glycosylated and non-glycosylated Asn-X-Thr/Ser acceptor sites: implications for protein engineering. Protein Eng 3: 433–442.

Gemmill, T.R., and Trimble, R.B. (1999) Overview of N- and O-linked oligosaccharide structures found in various yeast species. Biochim Biophys Acta 1426: 227–237.

Guan, Z., Naparstek, S., Kaminski, L., Konrad, Z., and Eichler, J. (2010) Distinct glycan-charged phosphodolichol carriers are required for the assembly of the pentasaccharide N-linked to the Haloferax volcanii S-layer glycoprotein. Mol Microbiol 78: 1294–1303.

Guan, Z., Meyer, B.H., Albers, S.V., and Eichler, J. (2011) The thermoacidophilic archaeon Sulfolobus acidocaldarius contains an unsually short, highly reduced dolichyl phosphate. Biochim Biophys Acta 1811: 607–616.

Guan, Z., Naparstek, S., Calo, D., and Eichler, J. (2012) Protein glycosylation as an adaptive response in Archaea: growth at different salt concentrations leads to

alterations in Haloferax volcanii S-layer glycoprotein N-glycosylation. Environ Microbiol 14: 743–573.

Hartman, A.L., Norais, C., Badger, J.H., Delmas, S., Haldenby, S., Madupu, R., et al. (2010) The complete genome sequence of Haloferax volcanii DS2, a model archaeon. PLoS One 5: e9605.

Helenius, J., Ng, D.T., Marolda, C.L., Walter, P., Valvano, M.A., and Aebi, M. (2002) Translocation of lipid-linked oligosaccharides across the ER membrane requires Rft1 protein. Nature 415: 447–450.

Helenius, A., and Aebi, M. (2004) Roles of N-linked glycans in the endoplasmic reticulum. Annu Rev Biochem 73: 1019–1049.

Huber, H., Hohn, M.J., Rachel, R., Fuchs, T., Wimmer, V.C., and Stetter, K.O. (2002) A new phylum of Archaea represented by a nanosized hyperthermophilic symbiont. Nature 417: 63–67.

Igura, M., Maita, N., Kamishikiryo, J., Yamada, M., Obita, T., Maenaka, K., and Kohda, D. (2008) Structure-guided identification of a new catalytic motif of oligosaccharyltransferase. EMBO J 27: 234–243.

Izquierdo, L., Schulz, B.L., Rodrigues, J.A., Güther, M.L., Procter, J.B., Barton, G.J., et al. (2009) Distinct donor and acceptor specificities of Trypanosoma brucei oligosaccharyltransferases. EMBO J 28: 2650–2661.

Izquierdo, L., Mehlert, A., and Ferguson, M.A. (2012) The lipid-linked oligosaccharide donor specificities of Trypanosoma brucei oligosaccharyltransferases. Glycobiology 22: 696–703.

Jarrell, K.F., and Albers, S.V. (2012) The archaellum: an old motility structure with a new name. Trends Microbiol 20: 307–312.

Jarrell, K.F., Ding, Y., Meyer, B.H., Albers, S.V., Kaminski, L., and Eichler, J. N- linked glycosylation in Archaea: A structural, functional and genetic analysis. Submitted.

Jones, G.M., Wu, J., Ding, Y., Uchida, K., Aizawa, S.I., Robotham, A., et al. (2012) Identification of genes involved in the acetamidino group modification of the flagellin N-linked glycan of Methanococcus maripaludis. J Bacteriol 194: 2693–2702.

Kaminski, L., Abu-Qarn, M., Guan, Z., Naparstek, S., Ventura, V.V, Raetz, C.R.H., et al. (2010) AglJ adds the first sugar of the N-linked pentasaccharide decorating the Haloferax volcanii S-layer glycoprotein. J Bacteriol 192: 5572–5579.

Kaminski, L., and Eichler, J. (2010) Identification of residues important for the activity of Haloferax volcanii AglD, a component of the Archaeal N-glycosylation pathway. Archaea 2010: 315108.

Kaminski, L., Guan, Z., Abu-Qarn, M., Konrad, Z., and Eichler, J. (2012) AglR is required for addition of the final mannose residue of the N-linked glycan decorating the Haloferax volcanii S-layer glycoprotein. Biochim Biophys Acta 1820: 1664–1670.

Kaminski, L., Lurie-Weinberger, M.N., Allers, T., Gophna, U., and Eichler, J. (2013b) Phylogenetic- and genome-derived insight into the evolution of N- glycosylation in Archaea. Mol Phylogenet Evol 68: 327–339.

Keenleyside, W.J., Clarke, A.J., and Whitfield, C. (2001) Identification of residues involved in catalytic activity of the inverting glycosyl transferase WbbE from Salmonella enterica serovar borreze. J Bacteriol 183: 77–85.

Kelly, J., Jarrell, H., Millar, L., Tessier, L., Fiori, L.M., Lau, P.C., et al. (2006) Biosynthesis of the N-linked glycan in Campylobacter jejuni and addition onto protein through block transfer. J Bacteriol 188: 2427–2434.

Lairson, L.L., Henrissat, B., Davies, G.J., and Withers, S.G. (2008) Glycosyltransferases: structures, functions, and mechanisms. Annu Rev Biochem 77: 521–555.

Larkin, A., Chang, M.M., Whitworth, G.E., and Imperiali, B. (2013) Biochemical evidence for an alternate pathway in N-linked glycoprotein biosynthesis. Nat Chem Biol 9: 367–373.

Lechner, J., Wieland, F., and Sumper, M. (1985a) Biosynthesis of sulfated saccharides N-glycosidically linked to the protein via glucose. Purification and identification of sulfated dolichyl monophosphoryl tetrasaccharides from halobacteria. J Biol Chem 260: 860–866.

Lechner, J., Wieland, F., and Sumper, M. (1985b) Transient methylation of dolichyl oligosaccharides is an obligatory step in halobacterial sulfated glycoprotein biosynthesis. J Biol Chem 260: 8984–8989.

Lechner, J., and Wieland, F. (1989) Structure and biosynthesis of prokaryotic glycoproteins. Annu Rev Biochem 58: 173–194.

Linton, D., Dorrell, N., Hitchen, P.G., Amber, S., Karlyshev, A.V, Morris, H.R., et al. (2005) Functional analysis of the Campylobacter jejuni N-linked protein glycosylation pathway. Mol Microbiol 55: 1695–1703.

Liu, D., Cole, R.A., and Reeves, P.R. (1996) An O-antigen processing function for Wzx (RfbX): a promising candidate for O-unit flippase. J Bacteriol 178: 2102–2107.

Magidovich, H., and Eichler, J. (2009) Glycosyltransferases and oligosaccharyltransferases in Archaea: putative components of the N-glycosylation pathway in the third domain of life. FEMS Microbiol Lett 300: 122–130.

Magidovich, H., Yurist-Doutsch, S., Konrad, Z., Ventura, V.V, Dell, A., Hitchen, P.G., and Eichler, J. (2010) AglP is a S-adenosyl-L-methionine-dependent

methyltransferase that participates in the N-glycosylation pathway of Haloferax volcanii. Mol Microbiol 76: 190–199.

Matsumoto, S., Igura, M., Nyirenda, J., Matsumoto, M., Yuzawa, S., Noda, N., et al. (2012) Crystal structure of the C-terminal globular domain of oligosaccharyltransferase from Archaeoglobus fulgidus at 1.75 A resolution. Biochemistry 51: 4157–4166.

Matsumoto, S., Shimada, A., and Kohda, D. (2013) Crystal structure of the C-terminal globular domain of the third paralog of the Archaeoglobus fulgidus oligosaccharyltransferases. BMC Struct Biol 13: 11.

Mescher, M.F., Strominger, J.L., and Watson, S.W. (1974) Protein and carbohydrate composition of the cell envelope of Halobacterium salinarium. J Bacteriol 120: 945– 954.

Mescher, M.F., and Strominger, J.L. (1976) Purification and characterization of a prokaryotic glucoprotein from the cell envelope of Halobacterium salinarium. J Biol Chem 251: 2005–2014.

Mescher, M.F., and Strominger, J.L. (1977) The shape-maintaining component of Halobacterium salinarium: a cell surface glycoprotein. Prog Clin Biol Res 17: 459– 465.

Mescher, M.F., and Strominger, J.L. (1978) Glycosylation of the surface glycoprotein of Halobacterium salinarium via a cyclic pathway of lipid-linked intermediates. FEBS Lett 89: 37–41.

Meyer, B.H., Zolghadr, B., Peyfoon, E., Pabst, M., Panico, M., Morris, H.R., et al. (2011) Sulfoquinovose synthase - an important enzyme in the N-glycosylation pathway of Sulfolobus acidocaldarius. Mol Microbiol 82: 1150–1163.

Meyer, B.H., Peyfoon, E., Dietrich, C., Hitchen, P., Panico, M., Morris, H.R., et al. (2013) Agl16, a thermophilic glycosyltransferase mediating the last step of N-Glycan biosynthesis in the thermoacidophilic crenarchaeon Sulfolobus acidocaldarius. J Bacteriol 195: 2177–2186.

Mullakhanbhai, M.F., and Larsen, H. (1975) Halobacterium volcanii spec. nov., a dead sea halobacterium with a moderate salt requirement. Arch Microbiol 104: 207– 214.

Namboori, S.C., and Graham, D.E. (2008) Acetamido sugar biosynthesis in the Euryarchaea. J Bacteriol 190: 2987–2996.

Naparstek, S., Vinagradov, E., and Eichler, J. (2010) Different glycosyltransferases are involved in lipid glycosylation and protein N-glycosylation in the halophilic archaeon Haloferax volcanii. Arch Microbiol 192: 581–584.

Nothaft, H., and Szymanski, C.M. (2010) Protein glycosylation in bacteria: sweeter than ever. Nat Rev Microbiol 8: 765–778.

Peyfoon, E., Meyer, B., Hitchen, P.G., Panico, M., Morris, H.R., Haslam, S.M., et al. (2010) The S-layer glycoprotein of the crenarchaeote Sulfolobus acidocaldarius is glycosylated at multiple sites with chitobiose-linked N-glycans. Archaea 2010: 754101.

Plavner, N., and Eichler, J. (2008) Defining the topology of the N-glycosylation pathway in the halophilic archaeon Haloferax volcanii. J Bacteriol 190: 8045–8052.

Prabakaran, S., Lippens, G., Steen, H., and Gunawardena, J. (2012) Post-translational modification: nature’s escape from genetic imprisonment and the basis for dynamic information encoding. Wiley Interdiscip Rev Syst Biol Med 4: 565–583.

Raetz, C.R.H., and Whitfield, C. (2002) Lipopolysaccharide endotoxins. Annu Rev Biochem 71: 635–700.

Rush, J.S., Gao, N., Lehrman, M.A., Matveev, S., and Waechter, C.J. (2009) Suppression of Rft1 expression does not impair the transbilayer movement of Man5GlcNAc2-P-P-dolichol in sealed microsomes from yeast. J Biol Chem 284: 19835–19842.

Sanyal, S., Frank, C.G., and Menon, A.K. (2008) Distinct flippases translocate glycerophospholipids and oligosaccharide diphosphate dolichols across the endoplasmic reticulum. Biochemistry 47: 7937–7946.

Sanyal, S., and Menon, A.K. (2009) Flipping lipids: why an’ what's the reason for? ACS Chem Biol 4: 895–909.

Sanyal, S., and Menon, A.K. (2010) Stereoselective transbilayer translocation of mannosyl phosphoryl dolichol by an endoplasmic reticulum flippase. Proc Natl Acad Sci U S A 107: 11289–11294.

Shams-Eldin, H., Chaban, B., Niehus, S., Schwarz, R.T., and Jarrell, K.F. (2008) Identification of the archaeal alg7 gene homolog (encoding N-acetylglucosamine-1- phosphate transferase) of the N-linked glycosylation system by cross-domain complementation in Saccharomyces cerevisiae. J Bacteriol 190: 2217–2220.

Silberstein, S., and Gilmore, R. (1996) Biochemistry, molecular biology, and genetics of the oligosaccharyltransferase. FASEB J 10: 849–858.

Sleytr, U.B., and Thorne, K.J. (1976) Chemical characterization of the regularly arranged surface layers of Clostridium thermosaccharolyticum and Clostridium thermohydrosulfuricum. J Bacteriol 126: 377–383.

Spiro, R.G. (2002) Protein glycosylation: nature, distribution, enzymatic formation, and disease implications of glycopeptide bonds. Glycobiology 12: 43R–56R.

Sumper, M., Berg, E., Mengele, R., and Strobel, I. (1990) Primary structure and glycosylation of the S-layer protein of Haloferax volcanii. 172: 7111–7118.

Szymanski, C.M., Yao, R., Ewing, C.P., Trust, T.J., and Guerry, P. (1999) Evidence for a system of general protein glycosylation in Campylobacter jejuni. Mol Microbiol 32: 1022–1030.

Szymanski, C.M., and Wren, B.W. (2005) Protein glycosylation in bacterial mucosal pathogens. Nat Rev Microbiol 3: 225–237.

VanDyke, D.J., Wu, J., Logan, S.M., Kelly, J.F., Mizuno, S., Aizawa, S.I., and Jarrell, K.F. (2009) Identification of genes involved in the assembly and attachment of a novel flagellin N-linked tetrasaccharide important for motility in the archaeon Methanococcus maripaludis. Mol Microbiol 72: 633–644.

Voisin, S., Houliston, R.S., Kelly, J., Brisson, J.R., Watson, D., Bardy, S.L., et al. (2005) Identification and characterization of the unique N-linked glycan common to the flagellins and S-layer glycoprotein of Methanococcus voltae. J Biol Chem 280: 16586–16593.

Wacker, M., Linton, D., Hitchen, P.G., Nita-Lazar, M., Haslam, S.M., North, S.J., et al. (2002) N-linked glycosylation in Campylobacter jejuni and its functional transfer into E. coli. Science 298: 1790–1793.

Waters, E., Hohn, M.J., Ahel, I., Graham, D.E., Adams, M.D., Barnstead, M., et al. (2003) The genome of Nanoarchaeum equitans: insights into early archaeal evolution and derived parasitism. Proc Natl Acad Sci U S A 100: 12984–12988.

Wieland, F., Dompert, W., Bernhardt, G., and Sumper, M. (1980) Halobacterial glycoprotein saccharides contain covalently linked sulphate. FEBS Lett 120: 110–114.

Wieland, F., Heitzer, R., and Schaefer, W. (1983) Asparaginylglucose: novel type of carbohydrate linkage. Proc Natl Acad Sci U S A 80: 5470–5474.

Wieland, F., Paul, G., and Sumper, M. (1985) Halobacterial flagellins are sulfated glycoproteins. J Biol Chem 260: 15180–15185.

Woese, C.R., and Fox, G.E. (1977) Phylogenetic structure of the prokaryotic domain: the primary kingdoms. Proc Natl Acad Sci U S A 74: 5088–5090.

Yan, A., and Lennarz, W.J. (2005) Two oligosaccharyl transferase complexes exist in yeast and associate with two different translocons. Glycobiology 15: 1407–1415.

Yao, Z.J., and Valvano, M.A. (1994) Genetic-analysis of the O-specific lipopolysaccharide biosynthesis region (rfb) of Escherichia coli K-12 W3110: identification of genes that confer group 6 specificity to Shigella flexneri serotype Y and 4a. J Bacteriol 176: 4133–4143.

Yurist-Doutsch, S., Abu-Qarn, M., Battaglia, F., Morris, H.R., Hitchen, P.G., Dell, A., and Eichler, J. (2008) AglF, aglG and aglI, novel members of a gene island involved in the N-glycosylation of the Haloferax volcanii S-layer glycoprotein. Mol Microbiol 69: 1234–1245.

Yurist-Doutsch, S., and Eichler, J. (2009) Manual annotation, transcriptional analysis, and protein expression studies reveal novel genes in the agl cluster responsible for N- glycosylation in the halophilic archaeon Haloferax volcanii. J Bacteriol 191: 3068– 3075.

Yurist-Doutsch, S., Magidovich, H., Ventura, V.V, Hitchen, P.G., Dell, A., and Eichler, J. (2010) N-glycosylation in Archaea: on the coordinated actions of Haloferax volcanii AglF and AglM. Mol Microbiol 75: 1047–1058.

Zähringer, U., Moll, H., Hettmann, T., Knirel, Y.A., and Schäfer, G. (2000) Cytochrome b558/566 from the archaeon Sulfolobus acidocaldarius has a unique Asn- linked highly branched hexasaccharide chain containing 6-sulfoquinovose. Eur J Biochem 267: 4144–4149.

תקציר

N-גליקוזילציה של חלבונים הינה אחת המודיפיקציות הכי שכיחות בטבע. אמנם בניגוד לאיפיון

המפורט של תהליך ה-N-גליקוזילציה באאוקריוטים ובחיידקים, הגרסה הארכאלית של תהליך זה נשארת

הכי פחות מובנת. בעשור האחרון הושקעו מאמצים בהבנת תהליך זה בארכאה ההלופילית Haloferax volcanii, על ידי התווית מסלול ה-archaeal glycosylation( Agl(. מסלול זה מעורב בהרכבה

ובהדבקה של פנטהסכריד המורכב מהקסוז, שתי חומצות הקסורונויות, מתיל אסתר של חומצה

הקסורונית ומנוז על שיירי החומצה האמינית אספרג'ין בעמדה 13 ובעמדה 83 של ה- S-layer glycoprotein, חלבון הדווח של תהליך ה-N-גליקוזילציה ב- Haloferax volcanii, כמו כן על שיירי

אספרג'ין של גליקופרוטאינים אחרים באורגניזם זה.

היום ידוע שב- Haloferax volcanii, תהליך ה-N-גליקוזילציה מערב הוספה של ארבעת

הסוכרים הראשונים של הפנטהסכריד בצורתם המשופעלת על גבי שייר ליפידי של דוליכול פוספט

)DolP( על ידי ארבע גליקוזילטראנספראזות באופן סדרתי. כתוצאה מהמחקר שלי AglJ זוהה

כגליקוזילטראנספראז הראשון, האחראי להוספה של ההקסוז הראשון על שייר של DolP. בהמשך ל-

AglJ גם AglG, AglI ו-AglE פועלים להוספה של שלושת הסוכרים הבאים על גבי שייר של DolP

המחובר כבר לסוכר הראשון. במקביל הגליקוזילטראנספראז AglD מוסיף מנוז, הסוכר האחרון של

הפנטהסכריד, על גבי שייר DolP ייחודי. אני הצלחתי לזהות את השיירים הקטליטיים של

גליקוזילטראנספראז זה. בהמשך להרכבה שלהם, DolP הקשור לטטרהסכריד ו- DolP הקשור למנוז

עוברים "פליפ" לצד החיצוני של הממברנה הפלסמטית. על ידי שיטות של ביולוגיה חישובית יחד עם

מחיקת גנים ואנליזות של ספקטרומטריית מסות של השייר הליפידי DolP ושל חלבון הדווח S-layer glycoprotein וסימון רדיואקטיבי של חלבון הדווח, הצלחתי להוכיח את התפקיד של AglR בתור

הפליפאז של DolP הקשור למנוז או בתור האנזים שמסייע לתהליך של הפליפ. לאחר שגם ה-DolP-

טטרהסכריד וגם DolP-מנוז עוברים לצד החיצוני של הממברנה, הטטרהסכריד מועבר לשיירי אספרג'ין

ע"י האוליגוסכרילטאנספראז AglB. ורק בשלב זה הסוכר האחרון, מנוז, מועבר מהשייר הליפידי שלו

לטטרהסכריד.

למרות שפגיעה או הפרעה של תהליך ה-N-גליקוזילציה פוגעת ביציבות ובמבנה של שכבת ה-S layer, וגם בעמידות של שכבה זו לחלבוני פרוטאזה, תהליך ה-N-גליקוזילציה אינו חיוני להישרדות של

Haloferax volcanii. יחד עם זאת, תהליך כזה שלאחר תרגום משתנה כתגובה לשינויים בתנאי הגידול.

במיוחד רואים זאת כאשר ה- S-layer glycoprotein מעוטר על ידי שני גליקנים שונים כתוצאה

משינויים במליחות הסביבה. שיירי אספרגי'ין בעמדה 13 ובעמדה 83 מעוטרים על ידי אותו הפנטהסכריד

שתיארתי מקודם כאשר תאים של Haloferax volcanii גדלים במדיום גידול המכיל 3.4 מולר של

סודיום כלוריד או 1.75 מולר של סודיום כלוריד, לעומת זאת, אספרג'ין בעמדה 498 מעוטר על ידי

גליקן טטרהסכרידי אחר המורכב מהקסוז מסולפט, שני הקסוזות ושייר של ראמנוז רק כאשר התאים

גדלים במליחות הנמוכה. אני אפיינתי את המרכיבים של מסלול חדש המעורב בבניה של הטטרהסכריד

הקשור לאספרג'ין בעמדה 498. מחקר זה מייצג את הדיווח הראשון על נוחכות של שני מסלולים של N-

גליקוזילציה המסוגלים יחדיו להיות מעורבים בשינוי של חלבון בודד באורגניזם. יחד עם זאת, למרות

הדיווח על הנוכחות של שני מסלולים של N-גליקוזילציה ב- Haloferax volcanii, רק

אוליגוסכרילסטראנספראז אחד, AglB, נמצא ב- Haloferax volcanii. ממצא זה הביא אותי לבחון את

הפיזור הפילוגנטי של תהליכי N-גליקוזילציה בענף הארכאלי, בהתבסס על הנוכחות או העדר של

.AglB

הצהרת תלמיד המחקר עם הגשת עבודת הדוקטור לשיפוט

אני החתום מטה מצהיר/ה בזאת: )אנא סמן(:

√ חיברתי את חיבורי בעצמי, להוציא עזרת ההדרכה שקיבלתי מאת מנחה/ים.

√ החומר המדעי הנכלל בעבודה זו הינו פרי מחקרי מתקופת היותי תלמיד/ת מחקר.

√ בעבודה נכלל חומר מחקרי שהוא פרי שיתוף עם אחרים, למעט עזרה טכנית הנהוגה בעבודה ניסיונית. לפי כך מצורפת בזאת הצהרה על תרומתי ותרומת שותפי למחקר, שאושרה על ידם ומוגשת בהסכמתם.

תאריך______שם התלמיד/ה לינה קמינסקי חתימה ______

העבודה נעשתה בהדרכת פרופ' ג'רי אייכלר

במחלקה למדעי החיים

בפקולטה למדעי הטבע

אנליזה של חלבוני Agl: מרכיבי תהליך ה-N-גליקוזילציה ב- Haloferax volcanii

מחקר לשם מילוי חלקי של הדרישות לקבלת תואר "דוקטור לפילוסופיה"

מאת לינה קמינסקי

הוגש לסינאט אוניברסיטת בן גוריון בנגב

אישור המנחה פרופ' ג'רי אייכלר______

אישור דיקן בית הספר ללימודי מחקר מתקדמים ע"ש קרייטמן______

כסלו, תשע"ד נובמבר, 3102

באר שבע

אנליזה של חלבוני Agl: מרכיבי תהליך ה-N-גליקוזילציה ב- Haloferax volcanii

מחקר לשם מילוי חלקי של הדרישות לקבלת תואר "דוקטור לפילוסופיה"

מאת לינה קמינסקי

הוגש לסינאט אוניברסיטת בן גוריון בנגב

כסלו, תשע"ד נובמבר, 3102

באר שבע