Research Collection

Doctoral Thesis

N-linked protein glycosylation: from eukaroytes to bacteria

Author(s): Wacker, Michael

Publication Date: 2002

Permanent Link: https://doi.org/10.3929/ethz-a-004394069

Rights / License: In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For moreinformation please consult the Terms of use.

ETH Library

N-linked protein glycosylation:

from eukaroytes to bacteria

A dissertation submitted to the

SWISS FEDERAL INSTITUTE OF TECHNOLOGY ZURICH

for the degree of

DOCTOR OF NATURAL SCIENCES

presented by

MICHAEL WACKER

Dipl. Natw. ETH

born on April 25, 1973

citizen of Beatenberg BE

accepted on the recommendation of

Prof. Dr. Markus Aebi, examiner

Prof. Dr. Linda Thöny-Meyer, co-examiner

Zurich 2002

Für meine Familie

1

Acknowledgment

Ich danke Prof. Markus Aebi für die ausgezeichnete Betreuung dieser Doktorarbeit.Er hat mir die Möglichkeit gegeben, ein ausserordentlich interessantes Gebiet kennenzu lernen und hat mich tatkräftig unterstützt.

Ich danke Prof. Linda Thöny-Meyer für ihr Interesse und für die Bereitschaft,Koreferentin dieser Dissertation zu sein.

Muchas gracias a Fabiana. Ella me soportó durante estos cuatro años, me ayudó consus ideas y su interés. Pero lo más importante es que me ayudó con su amor tangrande. Nunca paró de creer en mí y siempre me entendió. Le agradezco por su apoyoa pesar de estar tan lejos de su familia.

Vielen Dank an Peter, der mit seiner Diplomarbeit viel zum Gelingen dieser Arbeitbeigetragen hat und mir mit seiner offenen Art viel geholfen hat.

Ich danke Martin für seine Unterstützung. Er war mit seinem Fachwissenunverzichtbar und hat mir wesentlich bei der Reinigung der Glykoproteine geholfen.Daneben hat er auch im Fussballtraining immer in offenes Ohr für meine Problemegehabt, vielleicht hat er deshalb noch Ladehemmungen!

Thanks to Skip for his encouragement, for his suggestions, for critically reading mythesis and for many hours in famous restaurants and bars in Zurich.

Thanks to Mihai for his help during the last year, to Stephan for introducing me intoyeast genetics, to Karin and to all members of the yeast group as well as to the othermembers of the Aebi group.

Ich danke dem ganzen Institut, vor allem Palmira, Carmen, Paul, Alan und Jacquesfür ihren Einsatz.

Ein spezielles Dankeschön an Irene, Reka, Soco, Astrid, Pierce, Rinaldo, Alan,Mounir, Urs und Thomas für die vielen vergnüglichen Stunden am Institut und auchausserhalb...

Grazcha fich a mieus amihs da L’Engiadina: Sandro, Marco, Whisky, Fadri undGiandu. Ihr wart eine grosse Unterstützung! Vielen Dank auch an Nadia, Tanja,Natalia, Fränzi, Michi, Dave und meine Mitspieler vom FCA für die vielenvergnüglichen Stunden.

Meiner Familie danke ich für die riesige Unterstützung.

2

Table of contents

Summary 4

Zusammenfassung 6

1 Oligosaccharyltransferase of eukaryotes 8

1.1 Introduction 9

1.1.1 N-linked protein glycosylation pathway in yeast 9

1.1.2 The oligosaccharyltransferase catalyzes the central step in N-linked protein glycosylation

1.1.3 The oligosaccharyltransferase complex of Saccharomyces cerevisiae 12

1.2 Results 15

1.2.1 Two oligosaccharyltransferase complexes with different functions in vivo 15

1.3 Discussion 20

1.4 Materials and methods 23

1.4.1 Strains 23

1.4.2 Plasmids 23

1.4.3 Yeast manipulations 23

1.4.4 Preparation of microsomal membranes and solubilization of membrane proteins 23

1.4.5 Blue native polyacrylamide gel electrophoresis 24

1.4.6 Electroblotting of blue native gels and immunodetection 24

1.4.7 Oligosaccharyltransferase assay 25

1.4.8 Western blot analysis 26

2 Oligosaccharyltransferase of the bacterium Campylobacter jejuni 27

2.1 Introduction 28

2.1.1 N-linked protein glycosylation in archaea 29

2.1.2 Protein glycosylation in bacteria 30

2.1.3 The bacterium Campylobacter jejuni 31

2.1.4 Glycoproteins of C. jejuni 33

2.1.5 PglB, the putative oligosaccharyltransferase of C. jejuni 35

3

2.1.6 The pgl locus is essential for N-linked protein glycosylation in C. jejuni: a hypothesis 37

2.2 Results 39

2.2.1 Chemical deglycosylation affects protein antigenicity 39

2.2.2 Hydrophilic C-terminal domain of PglB is located in the periplasm 40

2.2.3 PglB plays a major role in protein glycosylation in C. jejuni 44

2.2.4 The locus with pglB mutations complements the pglE deletion 47

2.2.5 Purification and identification of two glycoproteins of C. jejuni 49

2.2.6 Deletion of acrA and ompH1 and complementation in trans 56

2.2.7 Expression of AcrA in E. coli in the presence of the pgl locus leads to glycosylation 58

2.3 Discussion 62

2.3.1 Pgl locus encodes N-linked protein glycosylation pathway of C. jejuni 62

2.3.2 Production of N-glycoproteins in E. coli 68

2.4 Material and methods 70

2.4.1 Bacterial strains and plasmids 70

2.4.2 Media and growth conditions 70

2.4.3 Protein electrophoresis and blotting 70

2.4.4 Chemical deglycosylation 71

2.4.5 Mutagenesis of C. jejuni genes 71

2.4.6 Construction of plasmids containing the pgl locus with mutations in the pglB locus 72

2.4.7 Conjugation of C. jejuni 74

2.4.8 Construction of plasmids encoding PglB-PhoA fusions 75

2.4.9 Alkaline phosphatase activity assay 76

2.4.10 Purification of C. jejuni glycoproteins and identification by mass spectrometry 77

2.4.11 Expression of AcrA and the pgl locus in E. coli 78

3 References 79

Curriculum vitae 90

4

Summary

N-linked protein glycosylation is an essential and highly conserved process in

eukaryotic cells. It starts with the stepwise assembly of an oligosaccharide on the lipid

dolichyl pyrophosphate at the cytoplasmic side of the membrane of the endoplasmic

reticulum (ER). The lipid-linked oligosaccharide (LLO) is flipped into the lumen of

the ER where full length oligosaccharide is obtained. In the central reaction of the

process, the oligosaccharyltransferase transfers the oligosaccharide

Glc3Man9GlcNAc2 to selected asparagine residues of proteins.

The oligosaccharyltransferase of Saccharomyces cerevisiae has been characterized in

detail. It is an enzyme complex consisting of nine highly conserved membrane

proteins: Stt3p (78 kDa), Ost1p (64 kDa), Wbp1p (45 kDa), Ost3p (34 kDa), Ost6p

(31 kDa), Swp1p (30 kDa), Ost2p (16 kDa), Ost5p (9 kDa) and Ost4p (3.4 kDa).

Yeast expresses two different OTase complexes differing only in the presence of

Ost3p and Ost6p, respectively. The components of the oligosaccharyltransferase

complex are highly conserved in eukaryotes.

In the first part of the thesis, it was shown that Ostp3 and Ost6p modulate the OTase

activity for the acceptor protein. Two membrane proteins were better substrates for

complex Ia (Ost3p containing complex) than for complex Ib (Ost6p containing

complex), whereas there was no difference in affinity for a soluble protein. A slightly

improved in vitro affinity for an acceptor peptide was observed for complex Ia,

confirming the results of the glycosylation studies in vivo.

Stt3p is the most highly conserved oligosaccharyltransferase subunit among different

species suggesting a direct role in N-linked protein glycosylation. Stt3p homologs are

also found in archaea where N-glycoproteins are produced. However the discovery of

a putative STT3 homolog in the genome of the Gram-negative bacterium

Campylobacter jejuni was surprising and suggested N-linked protein glycosylation in

this bacterium.

In the second part of the thesis, the role of the Stt3p homolog of C. jejuni, PglB, was

studied. It was shown that C. jejuni indeed is able to produce N-linked glycoproteins

and that PglB is directly involved in this process, most likely acting as an

oligosaccharyltransferase. Moreover, it became evident that the N-linked protein

glycosylation pathway of the bacterium shows significant similarities to the N-linked

5

protein glycosylation pathway of eukaryotes. Based on the sequence analysis of the

pgl locus of C. jejuni it is proposed that the oligosaccharide is assembled on a carrier

lipid at the cytoplasmic side of the plasma membrane, translocated to the periplasmic

space and transferred to selected proteins. Two N-glycoproteins of C. jejuni were

identified and the structure of the oligosaccharide was partially determined.

In the final part of the thesis, AcrA, a glycoprotein of C. jejuni, was expressed in

Escherichia coli in the presence of the C. jejuni pgl locus. Using this reconstituted

system it was possible to produce N-glycosylated AcrA protein in E. coli.

6

Zusammenfassung

Die N-gebundene Glykosylierung von Proteinen ist ein lebenswichtiger und hoch

konservierter Prozess in eukaryotischen Zellen. Er beginnt an der zytoplasmatischen

Oberfläche der Membran vom Endoplasmatischen Retikulum (ER) mit dem

schrittweisen Aufbau eines Oligosaccharids auf dem Lipid Dolichyl Pyrophosphat.

Das an das Lipid gebundene Oligosaccharid (LLO) wird dann in das Lumen des ER

transportiert, wo das LLO in seiner ganzen Länge erhalten wird. In der zentralen

Reaktion dieses Prozesses überträgt die Oligosaccharyltransferase das Oligosaccharid

Glc3Man9GlcNAc2 auf ausgewählte Asparginreste von Proteinen.

Die Oligosaccharyltransferase von Saccharomyces cerevisiae ist im Detail

charakterisiert. Sie ist ein Enzymkomplex aus neun hoch konservierten

Membranproteinen: Stt3p (78 kDa), Ost1p (64 kDa), Wbp1p (45 kDa), Ost3p (34

kDa), Ost6p (31 kDa), Swp1p (30 kDa), Ost2p (16 kDa), Ost5p (9 kDa) and Ost4p

(3.4 kDa). Hefe stellt zwei verschiedene Komplexe her, die sich nur in der

Anwesenheit von Ost3p oder Ost6p unterscheiden. Die Untereinheiten der

Oligosaccharyltransferase sind hoch konserviert in Eukaryoten.

Im ersten Teil der Dissertation wurde gezeigt, dass Ost3p und Ost6p die Aktivität für

den Proteinakzeptor anpassen. Zwei Membranproteine waren bessere Substrate für

Komplex Ia (Komplex, der Ost3p enthält) verglichen mit Komplex Ib (Komplex, der

Ost6p enthält). Hingegen wurde kein Unterschied in der Affinität für ein lösliches

Proteinsubstrat gefunden. Eine leicht verbesserte in vitro Affinität für ein

Peptidakzeptor wurde für Komplex Ia gemessen. Dies bestätigte die Resultate der

Glykosylierungsstudien in vivo.

Stt3p ist die am höchsten konservierte Untereinheit der Oligosaccharyltransferase in

verschieden Arten. Aufgrund dieser Tatsache wurde behauptet, dass Stt3p eine direkte

Rolle in der N-gebundenen Glykosylierung von Proteinen spielt. Homologe von Stt3p

wurden auch in Archaea gefunden, wo N-Glykoproteine produziert werden. Doch die

Tatsache, dass ein vermeintliches STT3 Homolog im Genom des Gram-negativen

Bakteriums Campylobacter jejuni gefunden wurde, war überraschend und liess

vermuten, dass N-gebundene Glykosylierung von Proteinen in diesem Bakterium

vorkommt.

7

Im zweiten Teil der Dissertation wurde die Rolle des Stt3p Homologs von C. jejuni,

PglB, studiert. Es wurde gezeigt, dass C. jejuni in der Tat fähig ist, N-gebundene

Glykoproteine zu produzieren und dass PglB direkt in diesen Prozess involviert ist,

am wahrscheinlichsten als Oligosaccharyltransferase. Ausserdem wurde klar, dass die

Biosynthese von N-gebundenen Glykoproteinen des Bakteriums wesentliche

Aenlichkeiten zur Biosynthese von N-gebundenen Glykoproteinen von Eukaryoten

zeigt. Basierend auf Sequenzähnlichkeiten des pgl Locus von C. jejuni wurde

vorgeschlagen, dass das Oligosaccharid auf dem Lipid an der zytoplasmatischen

Oberfläche der Plasmamembran synthetisiert wird, danach ins Periplasma

transportiert wird und auf ausgewählte Proteine übertragen wird. Zwei N-

Glykoproteine von C. jejuni wurden identifiziert und die Struktur des

Oligosaccharides teilweise aufgeklärt.

Im letzten Teil der Dissertation wurde AcrA, ein Glykoprotein von C. jejuni, in

Escherichia coli in Gegenwart des pgl Locus exprimiert. Durch Anwendung dieses

rekonstituierten Systems war es möglich, N-glykosyliertes AcrA Protein in E. coli zu

produzieren.

8

1 Oligosaccharyltransferase of eukaryotes

9

1.1 Introduction

1.1.1 N-linked protein glycosylation pathway in yeast

Protein glycosylation constitutes one of the most important post-translational protein

modifications in eukaryotic cells and may have numerous effects on physical

properties, structure, function and targeting of particular proteins. Especially the

carbohydrate moiety is to be regarded as having significant effects on both the

physicochemical features and on the structure of a protein, and it may affect its

enzymatic activity, antigenicity or thermal stability. In eukaryotes, carbohydrates are

predominantly linked to the protein either via the amide nitrogen of asparagine (N-

glycosidic bond) or the hydroxyl group of a serine or threonine (O-glycosidic bond)

residue.

N-linked protein glycosylation (for review see Helenius and Aebi 2001) is a common

protein modification found in eukaryotes and affects secretory proteins (Kornfeld and

Kornfeld 1985). The complex glycosylation process starts at the cytoplasmic face of

the endoplasmic reticulum (ER) with the assembly of an oligosaccharide on the

carrier lipid dolichyl pyrophosphate (Figure 1.1): two N-acetylglucosamine and five

mannose residues are attached to the lipid in a stepwise fashion, where nucleotide-

activated sugars serve as substrates. The lipid-linked oligosaccharide (LLO) is then

flipped into the lumen of the ER, a process catalyzed by the enzyme Rft1p (Helenius

et al., 2002). In the lumen of the ER, full length LLO is obtained by addition of

another four mannose and three glucose residues. In these reactions, Dol-P-Man and

Dol-P-Glc, respectively, are used as donor substrates. Finally, the transfer of the

oligosaccharide from the lipid to proteins is catalyzed by the oligosaccharyltransferase

(OTase) in the lumen of the ER.

The synthesis of the LLO and the transfer of the oligosaccharide to proteins is a

highly conserved process in eukaryotic cells. Using genetic and biochemical analysis,

most of the loci required for the synthesis of the lipid-linked oligosaccharide could be

identified in yeast. For most of the genes, homologs can be found in the genome of

eukaryotic cells (for review see Burda and Aebi 1999).

10

CTP CDP

SEC59

Lumen

DPM1

GDPGDPGDP-

ALG5 PP

P

P

Cytoplasm

ALG7

ALG1

ALG3 ALG9 ALG6 ALG8 ALG10ALG12

UDP

== =

=

STT3OST1WBP1OST3OST6SWP1OST2OST5OST4

dolichol

glucosemannoseGlcNAc

UDP

UDPUDP

UMP UDP

ALG2

ALG11

GDP

Fructose-6-PPMI40

SEC53

PSA1

PP

PPPP

MM

MM

M

MMMMMM

MMM

MM

ALG9

GG

MM

MM

MMMM

M

GGG

MMMM

MMMM

MMMMMMMM

MM

G

GGMM

GG

MM

MM

MMMM

M

GGG

MM

MM

MMMM

M

GGG

MMMM

MMMM

MMMMMMMM

MM

GGG

MM

MM

MMMM

M

GG

MMMM

MMMM

MMMMMMMM

MM

GMM

MM

MMMM

M

GMMMM

MMMM

MMMMMMMM

MM

MM

MM

MMMM

M

MMMM

MMMM

MMMMMMMM

MMM

MM

MMMM

MMM

MMMM

MMMMMMMM

MM

MM

MMMM

M

MMMM

MMMMMMMM

MMM

MMMM

MMM

MMMMMMMM

MMM

MMM

MMM

MMMMMM

MM

GDP MGDP MMM

M

MMM

MM

MMMM

M

MM

M

M

MMM

MM

M

-6-PM-6-PMM

MM

MM

PP PP

-1-PM-1-PMM

PPMM

PMPMM

UDP-GUDP-GG

GG

GG PP

CWH8

RFT1

5x

Figure 1.1: The N-linked protein glycosylation pathway in yeast. This highly conserved processstarts with the assembly of the oligosaccharide on the glycosyl carrier lipid dolichyl pyrophosphate atthe cytoplasmic side of the ER membrane. The LLO is flipped into the lumen of the ER, whereadditional mannose and glucose residues are added by specific glycosyltransferases. In the central stepof the reaction the oligosaccharyltransferase transfers the oligosaccharide from the carrier lipid dolichylpyrophosphate to proteins.

1.1.2 The oligosaccharyltransferase catalyzes the central step in

N-linked protein glycosylation

In the fundamental step of the N-linked protein glycosylation, the oligosaccharide is

transferred from the carrier lipid dolichyl pyrophosphate to the amide nitrogen of

selected asparagine residues of polypeptide chains. The acceptor polypeptide is

characterized by the consensus sequence Asn-X-Ser/Thr, where X can be any amino

acid except proline (Gavel and Von Heijne 1990). The OTase catalyzes this highly

complex reaction: thereby, the enzyme has to recognize the fully assembled LLO and

the polypeptide as the substrates and it has to catalyze the formation of the N-

glycosidic linkage. The full length LLO is the preferred substrate of the OTase, as the

glucose residues seem to enhance glycosyl transfer, presumably by leading to a

11

favorable conformation of the OTase that contributes to oligosaccharide recognition

and by affecting the apparent binding affinity for the acceptor substrate (Karaoglu et

al., 2001; Breuer and Bause 1995). It has been shown that the minimal glycosyl donor

in vitro is Dol-PP-GlcNAc2 (Sharma et al., 1981) and transfer of incomplete

oligosaccharides can be observed in vivo (Huffaker and Robbins 1983). In alg yeast

strains with a temperature sensitive block in LLO assembly or in alg mutants that

accumulate incompletely assembled LLO, shorter glycan chains are transferred to

proteins (Aebi et al., 1996; Burda et al., 1996; Burda and Aebi 1998; Burda et al.,

1999; Jackson et al., 1993; Reiss et al., 1996; Stagljar et al., 1994; te Heesen et al.,

1994). Genetic experiments in yeast revealed that the terminal α-1,2-linked glucose

residue is important for the substrate recognition by the OTase in vivo, because

mutants deficient in Alg10 activity hypoglycosylate proteins (Burda and Aebi 1998).

The linear tripeptide consensus sequence is not sufficient for the recognition by the

OTase because 10 - 30 % of potential sequons are either not glycosylated at all or are

not efficiently glycosylated (Gavel and Von Heijne 1990). Imperiali and coworkers

postulated that the Asn-X-Ser/Thr sequence has to adopt a specific conformation, the

Asn-turn, in order to be recognized as a peptide acceptor sequence (Imperiali et al.,

1992b) (Figure 1.2A). In vitro studies using synthetic peptide acceptor substrate

analog with a fixed Asn-turn motif as competitive inhibitor support this hypothesis

(Imperiali et al., 1992a). The Asn-turn possibly enhances proton dissociation from the

amide nitrogen generating an imidate and yielding the reactive nucleophile required

for the attack of the C-atom of the phosphoacetal-activated Dol-PP-oligosaccharide.

However, another model, proposed by Bause et al. (Bause and Legler 1981; Bause et

al., 1995; Bause et al., 1997; Breuer et al., 2001), favors the existence of a specific β-

turn where the β-amide nitrogen is the hydrogen-bond donor and the threonine/serine

hydroxyl group acts as an acceptor, resulting in enhanced β-amide nucleophilicity to

attack the C-1 atom of phosphoacetal-activated Dol-PP-oligosaccharide and

delivering the OH-proton to a base at the active site of the enzyme (Figure 1.2B).

Both models predict a specific three-dimensional conformation of the acceptor

sequence; however the experimental data available does not allow to exclude one of

the models.

12

B

A

NH

HN

O

O

O NH

H

N

O

H

O

HN

O

R

ENZ-Base

Dol-P-P-Oligosaccharide

H

NH

HN

O

O

O NH

H

N

O

H

O

HN

O

R

ENZ-Base

Dol-P-P-Oligosaccharide

H

NH

HN

O

O

N O

H

N

O

H

O

HN

O

R

H

H

ENZ-Base

NH

HN

O

O

N O

H

N

O

H

O

HN

O

R

H

Dol-P-P-Oligosaccharide

NH

HN

O

O

N O

H

N

O

H

O

HN

O

R

H

H

ENZ-Base

NH

HN

O

O

N O

H

N

O

H

O

HN

O

R

H

Dol-P-P-Oligosaccharide

Figure 1.2: Proposed mechanisms for OTase. Model A, postulated by Imperiali and coworkers(Imperiali et al., 1992b), assumes that an Asn-turn conformation for the peptide backbone is necessary.This turn is recognized by the OTase and allows the formation of a nucleophilic imidate bydelocalization of the electron pair and deprotonation. This molecule attacks the LLO, forming thesugar-protein bond. Model B, proposed by Bause and coworkers (Bause et al., 1997), prefers theexistence of a β-turn stabilized by the β-amide nitrogen and the hydroxyl group of the hydroxyl aminoacid.

1.1.3 The oligosaccharyltransferase complex of Saccharomyces

cerevisiae

The OTase enzyme is best studied in Saccharomyces cerevisiae, but comparative

analysis at a biochemical as well as at a genomic level suggests a high degree of

conservation of this enzyme in all eukaryotic cells (for review see (Knauer and Lehle

1999b; Silberstein and Gilmore 1996)). Whereas purification of active OTase from

ER membranes of higher eukaryotes showed that OTase activity is associated with a

heterotrimeric complex consisting of ribophorin I (66 kDa), ribophorin II (64 kDa),

and OST48 (48 kDa), purified yeast OTase consists of at least six different subunits

(Kelleher and Gilmore 1994). The yeast OTase complex is composed of Ost1p, a

protein homologous to ribophorin I of higher eukaryotic cells (Silberstein et al.,

13

1995b), Wbp1p (OST48) (Silberstein et al., 1992; te Heesen et al., 1992), Swp1p

(ribophorin II) (Kelleher and Gilmore 1994; te Heesen et al., 1993), Ost2p (DAD1)

(Silberstein et al., 1995a), Ost3p (Karaoglu et al., 1995) and Ost5p (Reiss et al.,

1997).

In addition to these six proteins, genetic screens have identified two other loci, OST4

and STT3 that are required for full OTase activity in vivo (Chi et al., 1996; Zufferey et

al., 1995). OST4 codes for a very small, 3.4 kDa hydrophobic protein. OST4 deleted

cells show a temperature-sensitive phenotype at 37 °C and a marked

hypoglycosylation of both soluble and membrane N-glycoproteins (Chi et al., 1996).

Additional genetic and biochemical evidence suggests that Ost4p is required for the

assembly of the OTase complex (Karaoglu et al., 1997; Spirig et al., 1997). The STT3

product is essential for vegetative growth of yeast cells and is highly conserved in

eukaryotes. Depletion of the protein leads to reduced OTase activity in vivo (Zufferey

et al., 1995). Recent experiments show that Stt3p is also a structural component of the

yeast OTase complex (Spirig et al., 1997; Karaoglu et al., 1997). Ost3p and Ost4p

have been proposed to act together in a subcomplex with the OTase subunit Stt3p

(Karaoglu et al., 1997; Knauer and Lehle 1999a; Spirig et al., 1997). Furthermore, the

overexpression of either Ost3p or Ost4p in defined stt3 mutant strains restores the

growth of the strains at 37 °C and improves glycosylation of carboxypeptidase Y

(CPY) (Spirig et al., 1997).

A search in the available databases for homologs of the non-essential Ost3p revealed

a hypothetical 37 kDa protein, now termed Ost6p, with 46% amino acid sequence

homology and 21% sequence identity to Ost3p (Knauer and Lehle 1999a). The

hydropathy plots of Ost3p and Ost6p are very similar, suggesting four potential

transmembrane domains and a similar predicted arrangement of an N-terminal signal

sequence (Karaoglu et al., 1997; Knauer and Lehle 1999a). It was shown that this

protein is indeed part of the OTase complex (Knauer and Lehle 1999a). Although

functional OTase complexes from S. cerevisiae have been isolated that do not contain

all the components described above but retain transferase activity in vitro (Knauer and

Lehle 1994; Pathak and Imperiali 1997), it was proposed that the OTase is a complex

consisting of nine subunits (Figure 1.3).

The OTase complex was analyzed by blue native polyacrylamide gel electrophoresis

that allows separation of proteins and protein complexes under native conditions

(Knauer and Lehle 1999a; Schägger and von Jagow 1991; Schägger et al., 1994;

14

Schägger 1995). It was shown that there were two different OTase complexes in yeast

only differing in the presence of Ost3p and Ost6p, respectively (Spirig 1999).

66kDa

669kDa

443kDa

132kDa

�S

tt3p

�O

st1p

�W

bp

1p

�S

wp

1p�

Ost

3p

�O

st6p

66kDa

669kDa

443kDa

132kDa

�S

tt3p

�O

st1p

�W

bp

1p

�S

wp

1p�

Ost

3p

�O

st6p

C

Ost1p(64 kDa)

N C

Ost5p(9 kDa)

C

N

Swp1p(30 kDa)

N

C

Wbp1p(45 kDa)

N

COstp2p(16 kDa)

C

N

Stt3p(78 kDa)

N C

Ost3p(34 kDa)

C

NOst4p(3.4 kDa)

N

cytoplasm

lumenN C

Ost6p(31 kDa)

Figure 1.3: OTase complex in yeast (adapted from Spirig 1999). Using blue native gelelectrophoresis and immunoblot with antisera against different subunits of the complex, it was shownthat the OTase in yeast is an oligomeric complex composed of at least eight different subunits. Allthese subunits are membrane-spanning proteins ranging from about 80 kDa to 3 kDa in molecularweight.

15

1.2 Results

1.2.1 Two oligosaccharyltransferase complexes with different

functions in vivo

It had been shown that there are two different OTase complexes in yeast (Spirig

1999). One form contains Ost3p and lacks Ost6p (termed complex Ia) and a second

form contains Ost6p and lacks Ost3p (complex Ib). Furthermore it had been

demonstrated that Ost4p is required for the stable integration of either of these

proteins into complex I (Spirig 1999). To test the hypothesis of two different OTase

complexes more directly, ∆ost3∆ost6 double mutant strains were generated and

transformed with a multicopy plasmid carrying either the OST3 or the OST6 locus.

According to the model, these strains should produce either complex Ia or complex

Ib. To visualize OTase, solubilized membrane proteins of these cells were separated

by blue native gel electrophoresis, transferred to PVDF membranes and the OTase

complexes were analyzed using αOst1p antibody (Figure 1.4). The ∆ost3∆ost6 double

mutant strains did not produce complex I and only low levels of complex II (lane 2),

whereas in both strains after overexpressing either Ost3p or Ost6p wild-type levels of

complex I were detected (lane 3 and 4, respectively). Complex II had been shown to

lack Ost4p and Ost3p/Ost6p. Therefore, comparable levels of either complex Ia (the

Ost3p containing complex) or complex Ib (the Ost6p containing complex) were

present in these two strains.

16

complex I

WT

Vec

tor

�o

st3�

ost

6 V

ecto

r

�o

st3�

ost

6 p

OS

T3

�o

st3�

ost

6 p

OS

T6

1 2 3 41 2 3 4

complex II

�Ost1p

Figure 1.4: Normal levels of OTase complex in engineered cells expressing only complex Ia orcomplex Ib. SS328 wild-type (WT, lane 1) and YG889 (∆ost3∆ost6) cells harboring either the vector(lane 2), the plasmid pOST3 (lane 3) or pOST6 (lane 4) were grown in liquid minimal medium lackinguracil at 23 ������������� �������������� ������������������������������������������������������������������������������������������������������� ���������������������� ��������!��"� ���������in a 5–12% gradient gel and blotted onto a PVDF membrane. OTase complexes (indicated by arrows)��������������� � #����������������

Next, the biogenesis of specific glycoproteins in theses strains was analyzed in order

to functionally differentiate the two OTase complexes (Figure 1.5). A strain lacking

both complex Ia and complex Ib expressed severely hypoglycosylated proteins, as

17

shown by the analysis of the glycoproteins CPY, Ost1p and Wbp1p (Figure 1.5, lane

2). It was concluded that complex II had reduced OTase activity in vivo. The strain

expressing only OTase complex Ia did not show any alterations of glycosylation

patterns with respect to these 3 proteins examined (Figure 1.5, lane 3), whereas

complex Ib showed a mild hypoglycosylation of Ost1p and Wbp1p (Figure 1.5, lane

4). These results suggested that both Ost3p and Ost6p were dispensable for OTase

function; however, their presence improved glycosylation efficiency in vivo.

Interestingly, both complex Ia and complex Ib were able to glycosylate CPY

efficiently (Figure 1.5, top panel, lane 3 and 4), whereas the two membrane proteins

Ost1p and Wbp1p were better substrates for the Ost3p-containing complex (Figure

1.5, middle and bottom panel, lane 3).

18

WT

Vec

tor

�o

st3�

ost

6 V

ecto

r

�o

st3�

ost

6 p

OS

T3

�o

st3�

ost

6 p

OS

T6

1 2 43

mCPY-1-2

Ost1p-1-2-3

Wbp1p-1-2

WT

Vec

tor

�o

st3�

ost

6 V

ecto

r

�o

st3�

ost

6 p

OS

T3

�o

st3�

ost

6 p

OS

T6

WT

Vec

tor

�o

st3�

ost

6 V

ecto

r

�o

st3�

ost

6 p

OS

T3

�o

st3�

ost

6 p

OS

T6

1 2 431 2 43

mCPY-1-2

Ost1p-1-2-3

Wbp1p-1-2

Figure 1.5: Reduced glycosylation efficiency in cells expressing OTase complex Ib. SS328 wild-type (WT, lane 1) and YG889 (∆ost3∆ost6) cells harboring either the vector YEp352 (lane 2), theplasmid pOST3 (lane 3) or pOST6 (lane 4) grown in liquid minimal medium lacking uracil at 30 °C tomid-log phase were analyzed. Protein extracts were prepared, separated by SDS-PAGE and Westernblot analysis was performed using CPY- (top), Ost1p- (middle) or Wbp1p- (bottom) specificantibodies. The positions of mature proteins (mCPY, Ost1p and Wbp1p) and the glycoforms lackingone (-1), two (-2) or three (-3) oligosaccharide moieties are indicated.

Our results suggest that the two OTase complexes differ in their affinity for specific

acceptor substrate in vivo. Therefore, the affinity of the different OTase complexes for

the acceptor peptide was measured in vitro. When OTase activity in extracts derived

19

from cells expressing only complex Ia, complex Ib or complex II was tested, a

difference in the Km values was observed: a slightly better affinity for the peptide was

observed for complex Ia as compared to complex Ib, confirming the results of the

glycosylation studies in vivo. For the mixture of the two complexes present in wild-

type derived extracts, an intermediate value for the apparent Km was determined.

Importantly, complex II had a higher Km than the fully assembled OTase (Table I).

This correlated with the strong hypoglycosylation of proteins in strains expressing

only complex II observed in vivo. Furthermore, it was concluded that

hypoglycosylation observed in strains expressing only complex Ib was due to the

reduced affinity of this OTase complex for the polypeptide substrate.

Strain:

(genotype)

SS328 +

YEp352

(WT)

YG889 + pOST3

( ���� ����

pOST3)

YG889 + pOST6

( ���� ����

pOST6)

YG889 + YEp352

( ���� ����)

Complex

formed:

Ia and Ib Ia Ib II

Km�$ %& 39 ± 12 40 ± 6 75 ± 17 106 ± 17

Table I: Km values of different OTase complexes for the acceptor peptide. The mean Km values ofOTase for the acceptor peptide in extracts derived from the strains indicated are given. Threeindependent measurements using two different extract preparations were performed. WT: wild-type.

20

1.3 Discussion

In recent years, not only in yeast but also in higher eukaryotes the OTase had been

shown to be composed of several different subunits (for review Knauer and Lehle

1999b; Silberstein and Gilmore 1996). Blue native gel electrophoresis allows the

visualization of proteins and protein complexes in their native form. It had been

successfully used for the analysis of OTase complexes present in different OTase

mutants (Spirig 1999). In wild-type extracts, one complex had been detected by

antibodies against different OTase subunits and had been termed complex I. A direct

correlation of the presence of complex I to OTase function was provided by the

analysis of different mutant strains with altered OTase activity: low levels of complex

I had been found in strains with strong hypoglycosylation of proteins, and the

reversion of such hypoglycosylation phenotypes by high copy number suppressors

had resulted in an increased appearance of complex I (Spirig 1999).

The combination of genetic approaches and the biochemical analysis by blue native

gel electrophoresis allowed the conclusion that two different OTase complexes exist

in yeast (Spirig 1999). These two complexes consist of at least 8 subunits each. They

differ only by the presence of Ost3p (complex Ia) and Ost6p (complex Ib),

respectively. Genetic and biochemical analysis suggest that the other proteins of the

complex Ia or Ib are Stt3p, Ost1p, Wbp1p, Swp1p, Ost2p, Ost5p and Ost4p (Spirig

1999). Ost3p and Ost6p share sequence similarity, and the similar hydropathy profiles

suggest a similar topology for the two proteins in the membrane (Knauer and Lehle

1999a). The existence of two distinct OTase complexes explains the observation that

most but not all of the cellular pool of Wbp1p, Swp1p, or Ost1p co-purified with

epitope-tagged Ost3p (Karaoglu et al., 1997). Also, the presence of a 31 kDa-band

which was speculated to represent YML019W (Ost6p) in purified OTase preparations

from strains expressing HA-tagged Stt3p but not from strains expressing HA-tagged

Ost3p (Karaoglu et al., 1997) supports the view that distinct OTase complexes exist in

yeast. Based on published results (Karaoglu et al., 1997) and on the analysis of native

OTase complexes, a four to one ratio of complex Ia to complex Ib can be assumed to

be present in vivo.

A set of large proteins as molecular size markers was used in blue native gel

electrophoresis. Such size markers had been reported to allow a reliable assessment of

the molecular masses of the protein import complexes of yeast mitochondria in blue

21

native gel electrophoresis (Dekker et al., 1996). Based on these markers, the mobility

of complex Ia or Ib suggested a mass of around 500 kDa. However, the expected

molecular mass of an OTase complex consisting of Stt3p, Ost1p, Wbp1p,

Ost3p/Ost6p, Swp1p, Ost2p, Ost5p, and Ost4p (Karaoglu et al., 1997; Kelleher and

Gilmore 1994; Spirig et al., 1997) is 280 kDa. Knauer and Lehle showed the

existence of an OTase complex with a size of 240 kDa by using blue native gel

electrophoresis (Knauer and Lehle 1999a). However they had used Nikkol to

solubilize the enzyme complex. Therefore, it is possible that active OTase was present

as a dimer using digitonin for solubilization of membrane proteins. Furthermore, since

not all the parameters that influence mobility of protein complexes in blue native gel

electrophoresis are known, it is premature to estimate the molecular mass of the

OTase complex based on its relative mobility in this gel system. Instead, size markers

served as standards for comparison of results of individual gels and not to deduce the

molecular mass of a given complex.

Genetic tools made it possible to generate yeast expressing only complex Ia or

complex Ib and allowed a functional assessment of the two complexes in vivo.

Analysis of native complexes in these strains confirmed that wild-type levels of fully

assembled OTase were present. Interestingly, the two complexes were functionally

different: in the absence of complex Ib, no hypoglycosylation of glycoproteins was

detected. In contrast, a mild hypoglycosylation was observed in strains expressing

only complex Ib. Additionally, complex Ib had a lower affinity for a hexapeptide

substrate in vitro. Probably Ost3p and Ost6p are involved in the recognition of

different polypeptide substrates. It was concluded that the hypoglycosylation observed

for Wbp1p and Ost1p was due to the reduced affinity of this OTase complex for the

polypeptide. However, complex Ia and Ib were able to glycosylate CPY efficiently,

but their presence was not required for OTase function. Strains lacking both Ost3p

and Ost6p were viable, hypoglycosylated proteins and expressed reduced levels of

complex II OTase. Ost3p and Ost6p might modulate the affinity of the OTase for

different acceptor polypeptides. Additional experiments will be required to address

the specific functions of these two complexes in vivo.

Previous experiments had suggested that both Ost3p and Ost6p are assembled at a late

time point in the assembly pathway of the OTase, because only assembled OTase

complex (Ia or Ib) containing either one of these subunits was observed (Spirig 1999).

Compatible with this observation is the hypothesis that Ost3p and Ost6p are

22

peripherally associated components of the OTase complex (Karaoglu et al., 1997).

For the stable association of both Ost3p and Ost6p, the small Ost4p subunit was

required. Therefore the Ost4p subunit seems to act as an assembly factor for

integration of Ost3p or Ost6p into the complex. This explains recent reports that

immunoprecipitation of Ost3p did not co-precipitate other OTase subunits in a ∆ost4

strain, whereas precipitation of Stt3p co-precipitated other OTase subunits except

Ost3p (Karaoglu et al., 1997). Whether the role in assembly of the OTase is the sole

function of Ost4p remains to be determined, but the lethal phenotype of ∆ost3∆ost4

double mutant strains had suggested that there were other roles of this protein besides

the integration of either Ost3p or Ost6p into the OTase complex (Spirig 1999).

In the past few years, much progress has been made in identifying the components of

the multimeric membrane complex OTase in yeast (Knauer and Lehle 1999b). The

functional significance of two distinct OTase complexes in yeast in vivo will be an

exciting line of future investigation. Biochemical analysis showed that chemical

modification of Wbp1p by methyl methanethiolsulfonate inhibited OTase activity in

vitro (Pathak et al., 1995). Ost1p was cross-linked to the peptide acceptor substrate

(Yan and Lennarz 1999). The mammalian homologs of these two proteins were also

cross-linked to substrate analogues (Bause et al., 1997), indicating a direct

involvement of these two components in catalysis. The results presented here provide

in vivo evidence that the Ost3p and Ost6p modulate the OTase activity for the

acceptor protein in vivo.

23

1.4 Materials and methods

1.4.1 Strains

SS328 (MATα ade2-101 ura3-52 his3∆200 lys2-801) (Vijayraghavan et al., 1989),

YG889 (MATa ade2-101 ura3-52 his3∆200 tyr1 ∆ost3::HIS3 ∆ost6::KanMX) (Spirig

1999).

1.4.2 Plasmids

pOST3 (Spirig et al., 1997) and pOST6 (Bättig 1998) were 2µ YEp 352 plasmids

containing the OST3 and OST6 gene, respectively.

1.4.3 Yeast manipulations

Standard yeast media and genetic techniques (Guthrie and Fink 1991) were used for

growth of yeast cultures.

1.4.4 Preparation of microsomal membranes and solubilization of

membrane proteins

Microsomal membranes were prepared as described (Reiss et al., 1997) with the

following modification: cells were grown as 2 l cultures and the membranes were

finally resuspended in 1 ml of membrane buffer (50 mM Tris/HCl (pH 7.4), 1 mM

MgCl2, 1 mM MnCl2, 35% glycerol) containing 1 mM DTT, 1 mM PMSF, and

protease inhibitors (aprotinin, antipain, chymostatin, leupeptin, and pepstatin, 2 µg/ml

each). To 100 µl of membrane suspension in membrane buffer 300 µl TM-buffer (50

mM Tris/HCl (pH 7.4), 0.2 M mannitol, 0.1 M NaCl, 1 mM MgCl2, 1 mM CaCl2, 1

mM MnCl2) containing 1 mM DTT, 1 mM PMSF, and protease inhibitors (as above)

was added. DNA was digested with 240 U DNAse I (3000 U/mg, Fluka, Buchs, CH)

for 45 min at 25 °C on a thermo shaker. Glycerol was added to a final concentration

of 10%, using 87% stock solution. Membrane proteins were solubilized by the

addition of digitonin (1.5% final concentration, 100 mg/ml stock solution, Sigma,

Buchs, CH) and 6-aminocaproic acid (750 mM final concentration, 0.5 mg/ml stock

solution, Fluka, Buchs, CH). Incubation was for 45 min at 4 °C with shaking.

Insoluble material was removed by centrifugation for 30 min at 100,000 x g at 4 °C.

Protein concentration of the collected supernatant was determined by the Bio-Rad

24

protein assay or with the method of Sailer and Weissmann (Sailer and Weissmann

1991) using bovine serum albumin as a standard. Samples were frozen in liquid

nitrogen and stored at -80 °C.

1.4.5 Blue native polyacrylamide gel electrophoresis

Blue native electrophoresis was carried out in a Bio-Rad Protean II cell (gel

dimensions: 20 x 15 x 0.15 cm). The gels consisted of a separating gel with a 5-12%

acrylamide gradient and a stacking gel (4% acrylamide). Buffers and gel compositions

were as described (Schägger and von Jagow 1991) except that Tris-Base at pH 7.5 (4

°C) was used in all buffers instead of Bistris at pH 7.0 (4 °C). Protein concentration of

the solubilized membrane protein samples was adjusted to 1 mg/ml with TM-buffer as

described above containing 750 mM 6-aminocaproic acid, 1.5% digitonin, 1 mM

DTT, 1 mM PMSF, 10% glycerol and protease inhibitors. Sample buffer (100 mM

Tris-HCl (pH 7.5 (4 °C)), 500 mM 6-aminocaproic acid, 5% Serva blue G) was added

(15% of the original sample volume), gently mixed and the sample was loaded on the

gel. The electrophoresis was carried out at 4 °C for 15 h with the current limited to 25

mA and the voltage limited to 380 V, and an additional hour with the current limited

to 25 mA and the voltage limited to 500 V. After 6 h, the cathode buffer (50 mM

Tricine, 15 mM Tris/HCl, 0.02% Serva blue G, pH 7.5(4 °C)) was removed and the

electrophoresis was continued with a cathode buffer containing no Serva blue G (50

mM Tricine, 15 mM Tris/HCl, pH 7.5(4 °C)).

1.4.6 Electroblotting of blue native gels and immunodetection

Gels were soaked for 5 min in transfer buffer (25 mM Tris-Base, 200 mM Glycine,

0.1% SDS, 20% methanol) and proteins were transferred onto PVDF membrane

(Milipore, Bedford, MA, USA) using a semi dry blotter from Kem-En-Tec

(Copenhagen, Denmark) at a constant current of 1 mA/cm2 for 135 min. PVDF

membranes were destained in 90% methanol. Blots were then air dried, blocked in

PBST (140 mM NaCl, 5 mM KCl, 10 mM Na2HPO4, 0.4 mM KH2PO4, 0.1% Tween

20) containing 10% milk powder and incubated with appropriate antisera. Antibody-

binding was visualized with peroxidase-labeled protein A using enhanced

Chemiluminescence (ECL) (Amersham, Little Chalfont, GB).

25

1.4.7 Oligosaccharyltransferase assay

Microsomal membrane proteins were prepared as described above and were adjusted

to 5 mg/ml with membrane buffer (50 mM Tris/HCl (pH 7.4), 1 mM MgCl2, 1 mM

MnCl2, 35% glycerol). Lipid-linked oligosaccharides (LLO) were extracted from 180

g of bovine pancreas as described (Spiro et al., 1976), dissolved in 40 ml

chloroform/methanol/water (10:10:3, v/v), separated from insoluble material by

centrifugation at 4,000 x g at room temperature and stored at -20 °C. The DEAE

purification step was omitted. The terminally acetylated and amidated hexapeptide

Nα-Ac-YNLTSV-NH2 (Wieland et al., 1987) was purchased from Tana Laboratories

(Houston, TX, USA) and iodinated as described (Kelleher et al., 1992) except that the

amount of chloramine T was doubled. The iodinated hexapeptide was dissolved in

'%(#������������������""� %������������)�*��

OTase activity was assayed as following (Kelleher et al., 1992; Das and Heath 1980):

��� � �!� ++#� $"�)�� %� !���� �������������&������ ������ ��� �� (������ (�������� ���

��������������"� �,,��%�-���.��$�.�/��&��00��%�1����)��%�%��2, 185 mM

sucrose, 0.3% NP-40, 1 mM DTT. Insoluble material was removed by centrifugation

���� 0"� � ����������� $� ���� � 230""� %� !���� ��������������� ������� ��� � ���

����������� ��� �"� � /"""� ���4���&�� - �� ��������� ���� �������� � � ������� 2"�

����������������� $�""� ���������&� ��������������!!����5!���� ��������������20� *�

!���0"������ � �� ������������������������ ����� � � ������������!��""� �1�)"� $2�&�

Glycosylated peptide was recovered by adding 1 ml ice-cold wash buffer (50 mM

Tris-HCl (pH 7.4), 1 M NaCl, 1 mM MnCl2, 1 mM MgCl2, 1 mM CaCl2, 0.01% NP-

)"&� ��� � �� ���� � �������� !������ � � �""� � ������������ 5(�� ������ �����

(Pharmacia, freshly suspended in wash buffer). The tubes were incubated for 20 min

on a rotating wheel at 4 °C and the beads were washed three times with 1 ml wash

buffer. The radioactivity retained on the beads was quantified using a COBRA II

(Canberra Packard) auto-gamma counter. Heat inactivated membranes served as a

control.

To determine the Km value for the hexapeptide substrate, the final substrate

�������������� ���� ������� �������� 2� %� ���� 0""� %�� 6��� ��� � ���������

concentration, triplicate assays were performed.

26

1.4.8 Western blot analysis

Western blot analysis of different glycoproteins has been described (Burda et al.,

1996).

27

2 Oligosaccharyltransferase of the bacterium

Campylobacter jejuni

28

2.1 Introduction

N-linked protein glycosylation in the endoplasmic reticulum (ER) follows a complex,

but highly conserved pathway in all eukaryotic cells. The initial step of this pathway

is the assembly of a branched oligosaccharide on the carrier lipid dolichyl

pyrophosphate at the membrane of the ER. In the central reaction of the process, this

lipid-linked oligosaccharide is transferred to selected asparagine residues of newly

synthesized polypeptides in the lumen of the ER catalyzed by the

oligosaccharyltransferase (OTase). In this multisubunit enzyme, the essential Stt3

subunit is the most highly conserved among different species suggesting a direct role

in N-linked protein glycosylation (Figure 2.1).

Pyrococcus abyssi

Pyrococcus horikoshiiMethanobacteriumthermoautotrophicum

Methanococcus jannaschii

Archaeoglobus fulgidus(S-layer glycoproteins)

Leishmania majorToxoplasma gondii

Drosophila melanogasterSaccharomycescerevisiae

Homo sapiensMus musculus

Pyrococcus horikoshii

Pyrococcus abyssi

Archaea

Campylobacter jejuniPglB BacteriaCampylobacter jejuniPglB Bacteria

Eukarya

Figure 2.1: Phylogenetic tree of homologs of Stt3p. Stt3p is the most highly conserved subunit of theOTase. There are homologs found in all eukaryotes as well as in archaea. N-glycoproteins play anessential role in these organisms. Interestingly there is also a homolog of Stt3p, PglB, encoded by agene of the bacterium Campylobacter jejuni.

Especially the predicted topological orientation of the Stt3 proteins is highly

conserved among species. The N-terminal domain, 2/3 of the protein, is highly

29

hydrophobic whereas the C-terminal domain is highly hydrophilic. The latter domain

also contains the most conserved amino acid sequence. This domain of the protein

was shown to be localized in the lumen of the ER in case of the yeast protein

(Zufferey et al., 1995). N-glycoproteins are produced in archaea and it has been

postulated that the Stt3p homologs found in archaea act as OTase (Spirig 1999).

However the discovery of an STT3 homolog, pglB, in the genome of C. jejuni came as

a surprise and suggested N-linked protein glycosylation in this bacterium. Until now it

has been the only homolog found in bacteria that are thought to be unable to perform

N-linked protein glycosylation.

2.1.1 N-linked protein glycosylation in archaea

N-linked protein glycosylation is not restricted to eukaryotes but also occurs in

archaea. Detailed studies with Halobacterium halobium showed that the S-layer

protein, the major cell wall component, contains two different N-linked

oligosaccharides (Lechner and Wieland 1989). The similarity of the glycosylation

process in H. halobium to the eukaryotic system is evident (Figure 2.2). As in

eukaryotes, the sequences that are glycosylated are characterized by the Asn-X-

Ser/Thr consensus motif and the covalent linkage of the oligosaccharide occurs via

the amide nitrogen group of the Asn residue. Two different N-linked oligosaccharides

are found on the protein: (1) one glycosaminoglycan chain, constructed by a repeating

sulfated pentasaccharide block is linked to the protein via the asparaginyl-GalNAc

and (2) ten sulfated oligosaccharides that contain glucose, glucuronic acid, and

iduronic acid are bound to the protein via asparaginylglucose. They are synthesized

on the carrier lipid dolichyl pyrophosphate and dolichyl phosphate at the cytoplasmic

side of the membrane, translocated to the outside and in the central reaction of the

process, the oligosaccharide is transferred to the S-layer protein (Figure 2.2) (Lechner

and Wieland 1989). Interestingly, not only two different oligosaccharides are

synthesized on the carrier lipid and transferred to the S-layer protein, but also

different STT3 homologs are found in the genome of the archaeal species sequenced

so far. It is interesting to note that one of the STT3 homologous ORF of

Archaeoglobus fulgidus is located within an operon that encodes many of the

functions required for biosynthesis of a N-linked oligosaccharide according to the

model mentioned above (Burda and Aebi 1999) and has supported the idea that the

Stt3 proteins are oligosaccharyltransferases. In addition, this hypothesis points to an

30

archaeal origin of N-linked protein glycosylation. In archaea, N-linked sugars serve

primarily as structural components of the cell wall, whereas in eukaryotic cells N-

glycans serve several different functions (Helenius and Aebi 2001).

GLC-(HEXUA)3-GLC-P-DOL

me-GLC-(HEXUA)3-GLC-P-DOL

DOL-P-GLC-(HEXUA)3-GLC-me

Thr/Ser-X-Asn

N-Ala-Asn-Ala-Ser

DOL-P-P- 10DOL-P-P- 10

C

inside outside

me-OH

Figure 2.2: Hypothesis of the synthesis of N-glycoproteins in Halobacterium halobium (adaptedfrom Lechner and Wieland 1989). The repeating block of a sulfated pentasaccharide is assembled onthe carrier lipid dolichyl pyrophosphate and the sulfated oligosaccharide is assembled on dolichylphosphate at the cytoplasmic side of the membrane. The lipid-linked oligosaccharide is translocatedacross the membrane and transferred to the consensus sequence Asn-X-Ser/Thr of the S-layer protein.This central reaction is proposed to be catalyzed by two different Stt3 proteins. The GalNAc of theglycosyminoglycan is linked to the amide nitrogen of the Asn within the sequence Asn-Ala-Ser. TheSer is not essential for glycosylation proposing that there is a novel type of N-glycosyltransferase(Zeitler et al., 1998).

2.1.2 Protein glycosylation in bacteria

Although glycosylation of proteins was once thought to be limited to eukaryotes and

S-layer proteins of archaea (Schäffer and Messner 2001), enzymes involved in sugar

polymer degradation, subunits of flagellins, pilins and fimbriae in different bacterial

31

species have been reported to be glycosylated (reviewed in Messner 1997; Moens and

Vanderleyden 1997; Schäffer et al., 2001). Thus, the ability to glycosylate proteins is

ubiquitous in all kingdoms of life. Furthermore it became evident that the

glycosylation process in prokaryotes leads to a much greater diversity of glycan

compositions, linkage units (amino acid–attached sugar), and glycosylation sequences

on polypeptides than in eukaryotic glycoproteins. Due to the abundance of

polysaccharides associated with the outer surface of bacteria, extensive purification of

glycoproteins is needed to separate them from the lipopolysaccharides (LPS) and

extracellular polysaccharides (EPS). In most of these glycoproteins, the sugars were

shown to be O-linked.

It is proposed that bacterial glycoproteins are responsible to maintain protein

conformation and stability, to protect against proteolytic degradation and to be

involved in surface or intracellular recognition and cell adhesion. However the

experimental evidence is scarce. Important pathogens, such as Neisseria,

Mycobacterium, Streptococcus and Campylobacter species, synthesize glycoproteins

that are involved in crucial physiological processes and play a critical role in

pathogenicity (Stimson et al., 1995; Dobos et al., 1996; Erickson and Herzberg 1993;

Szymanski et al., 1999).

2.1.3 The bacterium Campylobacter jejuni

Campylobacter jejuni is an important human pathogen, causing acute human

enterocolitis, and it is the most significant cause of food-borne diarrhea in many

industrialized countries (Ketley 1997; van Vliet and Ketley 2001). Disease outcomes

from C. jejuni infection vary from mild, noninflammatory, self-limiting diarrhea to

severe, inflammatory, bloody diarrhea lasting for several weeks. In addition, C. jejuni

is also associated with the development of peripheral neuropathies, Miller-Fisher and

Guillain-Barré syndromes (Nachamkin et al., 1998). Despite its importance as a

human pathogen, the understanding of the mechanisms of Campylobacter-associated

disease is still relatively poor. Members of the Campylobacter species colonize the

gastrointestinal tract of a broad range of animals. They are commensal in most

animals, but notably are associated with disease in humans.

In vitro studies with several cell lines of intestinal origin showed that colonization of

the intestine requires the ability to move into the mucus layer that covers the intestinal

cells. The motility is conferred by the polar flagella of C. jejuni and combined with its

32

cork-screw form it allows the bacterium to efficiently penetrate this mucus barrier.

Besides motility, chemotaxis is essential for C. jejuni colonization. An important

feature in C. jejuni pathogenesis is its binding and entry into host cells (Wooldridge

and Ketley 1997). Upon infection, C. jejuni crosses the mucus layer covering the

epithelial cells and adheres to these cells. A subpopulation of these cells subsequently

invades the epithelial cells. Invasion then triggers inflammation and, together with the

production of toxins, this can lead to damaging the epithelial cell function. As a result,

the normal absorptive capacity of the intestine is perturbed.

C. jejuni is a microaerophilic organism; it has to be cultured in an atmosphere

containing 5–10% oxygen and 5–10% carbon dioxide. C. jejuni is naturally

transformable, and conjugation has also been used for the introduction of DNA

(Guerry et al., 1994). Plasmids that do not contain an origin of replication of

Campylobacter, but contain Campylobacter-derived antibiotic resistance genes, are

used for deletion of genes in C. jejuni by homologous recombination. Mutants can be

complemented in trans by introduction of a C. jejuni - E. coli shuttle plasmid through

conjugation. These plasmids contain both Escherichia and Campylobacter origins of

replication and Campylobacter-derived antibiotic resistance genes. However,

transposons of either Gram-negative or Gram-positive origin have not been found to

transpose in Campylobacter. Due to this lack of traditional genetic analysis and the

lack of suitable animal models to assess virulence, the pathophysiology of C. jejuni is

poorly understood (for review see Newell 2001).

The C. jejuni genome was sequenced and contains a small circular chromosome of

around 1.6 MBp of AT-rich DNA (Parkhill et al., 2000). The low GC content (around

30 %) can cause problems when cloning these A + T rich sequences in E. coli. The

small size of the genome is probably reflected in the requirement of Campylobacter

for complex media for growth and their inability to ferment carbohydrates. There is

almost a complete lack of repetitive DNA sequences and 94% of the genome code for

1654 proteins, making it the densest bacterial genome sequenced to date. One of the

most striking findings is the presence of hypervariable sequences. These are mainly

length variations in poly G:C tracts found in genes encoding the biosynthesis or

modification of surface structures. This variation in length may be produced by

slipped-strand mispairing during replication. It affects translation as has been shown

for phase variation of surface properties or antigenicity. Completion of the C. jejuni

33

NCTC 11168 genome sequence offers opportunities to understand the molecular basis

of virulence of this major pathogen.

2.1.4 Glycoproteins of C. jejuni

The enteric pathogen C. jejuni has been known to contain glycoproteins (Alm et al.,

1992). Flagellins from C. jejuni are glycosylated (Doig et al., 1996). Although

different findings had pointed to sialic acid (N-acetylneuramic acid) as a modification

of flagellins (Doig et al., 1996; Guerry et al., 1996; Linton et al., 2000), the glycosyl

component was identified recently as pseudoaminic acid, an unusual sugar that is

related to sialic acid (Thibault et al., 2001). The genes involved in flagellin

glycosylation are located in the same gene cluster as the ORF encoding the structural

flagellin subunits (flaA and flaB), defining the flagellin biosynthesis and glycosylation

locus (Guerry et al., 1996; Linton et al., 2000; Thibault et al., 2001).

Recently, two other glycoproteins of C. jejuni NCTC 11168 have been purified and

identified (Linton et al., 2002). PEB3 is known as a highly immunogenic, acid-

extractable protein of C. jejuni (Pei et al., 1991). The second glycoprotein identified

was called CgpA (Campylobacter glycoprotein), a putative periplasmic protein.

Linton and colleagues showed that the pgl locus is required for the glycosylation of

these two proteins (Linton et al., 2002).

Genes present in the pgl locus encode proteins with significant sequence similarity to

enzymes involved in bacterial lipopolysaccharide (LPS) and capsular polysaccharide

biosynthesis (Table I). This locus was also identified in two different C. jejuni strains

(Wood et al., 1999; Fry et al., 1998) and was named wla based on the nomenclature

proposed for LPS associated genes (Reeves et al., 1996). Interestingly mutagenesis of

some genes of this locus had no effect on LPS biosynthesis but instead caused a

dramatic reduction in the immunoreactivity of various proteins (Szymanski et al.,

1999). Furthermore, Szymanski and coworkers showed that chemical deglycosylation

of Campylobacter proteins also led to a reduction in immunoreactivity. Therefore the

locus was renamed pgl (for protein glycosylation) using the nomenclature based on a

locus involved in pilin glycosylation of Neisseria meningitidis (Jennings et al., 1998).

Hence, the pgl nomenclature will be used in this thesis.

C. jejuni

protein

Similar

protein

Proposed function GenBank

accession

34

number

GalE

WlaB

PglH

PglI

PglJ

PglB

PglA

PglC

PglD

PglE

PglF

PglG

RfaC

MsbA

TrsD

TrsB

TrsE

Stt3p

PglA

WbaP

PglB

PglC

PglD

AcfB

Escherichia coli

Escherichia coli

Yersinia enterocolitica

Yersinia enterocolitica

Yersinia enterocolitica

Saccharomyces cerevisiae

Neisseria meningitidis

Salmonella enterica

Neisseria meningitidis

Neisseria meningitidis

Neisseria meningitidis

Vibrio cholerae

UDPglucose 4-epimerase

Export of lipid A core

Glycosyltransferase

Glycosyltransferase

Glycosyltransferase

Oligosaccharyltransferase

Galactosyltransferase (Pilin glycosylation)

Galactosyltransferase first step

Acetyltransferase (Pilin glycosylation)

Perosamine synthetase (Pilin glycosylation)

Dehydratase (Pilin glycosylation)

Accessory colonization factor

P09147

P27299

S51263

S51261

S51264

JC4355

AAC35426

P264006

AAC25979

AAC25980

AAC25981

P06657

Table I: pgl locus and proposed function of genes. Based on homologies to known genes, functionsof the single genes were proposed and a hypothesis developed (see figure 2.4) (adapted from Fry et al.,1998).

35

2.1.5 PglB, the putative oligosaccharyltransferase of C. jejuni

PglB, a protein encoded by one gene of the pgl locus, exhibits homologies to Stt3p

(see Figure 2.1 and Table I). Until now it is the only homolog of the OTase found in

bacteria. Figure 2.3 shows a hydropathy analysis of the amino acid sequence of PglB

and Stt3p. As already mentioned, the predicted membrane topology is highly

conserved. 2/3 of the polypeptide, located at the N-terminus, is hydrophobic and this

domain is proposed to contain 10 to 12 transmembrane domains. In contrast, the C-

terminal domain is highly hydrophilic and contains the WWDYGY motif, highly

conserved among the Stt3 protein family. In yeast, it was shown that this hydrophilic

C-terminus of Stt3p is located in the lumen of the ER, where the transfer of the

oligosaccharide to the protein takes place (Zufferey et al., 1995).

36

100 200 300 400 500 600 700

-1.0

0.01.02.0

Residue

Kyte-Doolittle hydropathy: Stt3p

100 200 300 400 500 600 700-1.0

0.0

1.0

2.0

ResidueKyte-Doolittle hydropathy: PglB

M L K K E Y L K - - - - - - N P Y L V L F A M I V L A Y V F - - S V F C R F Y W V W W - - - - A S E F N E Y F - F N N Q L M I I S N D G Y A F A E - - G A R D - 1 Cj

M T S T T A A R T A S S R V G A - T T L L T I V V L A L A W F V G F A S R L F A I V R F E S I I H E F D P W F N Y R A T H H M V Q H G F Y K F L N W F D E R A W 1 Ce

M N R T - - P K M L N S K V A G Y S S L I T F A I L L I A W L A G F S S R L F A V I R F E S I I H E F D P W F N Y R A T A Y M V Q N G W Y N F L N W F D E R A W 1 Dm

M T K F G F L R L S Y E K Q D - - - T L L K L L I L S M A A V L S F S T R L F A V L R F E S V I H E F D P Y F N Y R T T R F L A E E G F Y K F H N W F D D R A W 1 Hs

M G S D - - - - - R S C V L S V F Q T I L K L V I F V A I F G A A I S S R L F A V I K F E S I I H E F D P W F N Y R A T K Y L V N N S F Y K F L N W F D D R T W 1 Sc

- - - - - M I A G F H Q P N D L S Y Y G S S L S T L T Y W L Y K I T P F S F E S I I L Y M S T F L S S L V V I P I I L L A N E Y K R P L M G F V A A L L A S V A 65 Cj

Y P L G R I V G G T V Y P G L M V T S G L I H W I L D S - L - - N F H V H I R E V C V F L A P T F S G L T A I A T Y L L T K E L W S P G A G L F A A C F I A I S 80 Ce

Y P L G R I V G G T V Y P G L M I T S G G I H W L L H V - L - - N I P V H I R D I C V F L A P I F S G L T S I S T Y L L T K E L W S A G A G L F A A S F I A I V 79 Dm

Y P L G R I I G G T I Y P G L M I T S A A I Y H V L H F - F - - H I T I D I R N V C V F L A P L F S S F T T I V T Y H L T K E L K D A G A G L L A A A M I A V V 78 Hs

Y P L G R V T G G T L Y P G L M T T S A F I W H A L R N W L - - G L P I D I R N V C V L F A P L F S G V T A W A T Y E F T K E I K D A S A G L L A A G F I A I V 76 Sc

N S Y Y N R T M S G Y Y D T D M L V I V L P M F I L F F M V R M I L K K D F F - - S L I A L P L F I G I Y L W W Y P S S Y T L N V A L I G L F L I Y T L I F H R 140 Cj

P G Y T S R S V A G S Y D N E G I A I F A L Q F T Y Y L W V K S L K T G S I M W A S L C A L S Y F Y M V S A W - - - G G Y V F I I N L I P L H A L A L I I M G R 157 Ce

P G Y I S R S V A G S Y D N E G I A I F A L Q F T Y F L W V R S V K T G S V F W S A A A A L S Y F Y M V S A W - - - G G Y V F I I N L I P L H V F V L L I M G R 156 Dm

P G Y I S R S V A G S Y D N E G I A I F C M L L T Y Y M W I K A V K T G S I C W A A K C A L A Y F Y M V S S W - - - G G Y V F L I N L I P L H V L V L M L T G R 155 Hs

P G Y I S R S V A G S Y D N E A I A I T L L M V T F M F W I K A Q K T G S I M H A T C A A L F Y F Y M V S A W - - - G G Y V F I T N L I P L H V F L L I L M G R 154 Sc

- K E K I F - - - - - - Y I A V I L S S L T L S N I A W F - Y Q S A I I V I L F A L F A L E Q - - - - - - - - - - - - K R L N F M I I - - - - - - - - - - G I L 218 Cj

Y S S R L F V S Y T S F Y C L A T I L S M Q V P F V G F Q P V R T S E H M P A F G V F G L L Q I V A L M H Y A R N R I T R Q Q F M T L F V - G G L T I - L G A L 234 Ce

Y S P R L L T S Y S T F Y I L G L L F S M Q I P F V G F Q P I R T S E H M A A L G V F V L L M A V A T L R H L Q S V L S R N E F R K L F I V G G L L V G V G V F 233 Dm

F S H R I Y V A Y C T V Y C L G T I L S M Q I S F V G F Q P V L S S E H M A A F G V F G L C Q I H A F V D Y L R S K L N P Q Q F E V L F R S V I S L V G F V L L 232 Hs

Y S S K L Y S A Y T T W Y A I G T V A S M Q I P F V G F L P I R S N D H M A A L G V F G L I Q I V A F G D F V K G Q I S T A K F K V I M M V S L F L I - - L V L 231 Sc

G S A T L I F L I L S G G V D P I L Y Q L K F Y I F R S D E S A N L T Q G F M Y F N V N Q T I Q E V E N V D F S E F M R R I S G S E I V F L F S L F G F V W L L 268 Cj

S V V V Y F A L V W G G Y V A P - - F S G R F Y S L W D T G Y A K - - - - - I H I P I I A S V S E H Q P T T W V S F F F D L H I T A A V F P V G L W Y C I K K V 312 Ce

V A V V V L T M L - - G V V A P - - W S G R F Y S L W D T G Y A K - - - - - I H I P I I A S V S E H Q P T T W F S F F F D L H I L V C A F P V G V W Y C I K Q I 313 Dm

T V G A L L M L - - T G K I S P - - W T G R F Y S L L D P S Y A K - - - - - N N I P I I A S V S E H Q P T T W S S Y Y F D L Q L L V F M F P V G L Y Y C F S N L 312 Hs

G V V G L S A L T Y M G L I A P - - W T G R F Y S L W D T N Y A K - - - - - I H I P I I A S V S E H Q P V S W P A F F F D T H F L I W L F P A G V F L L F L D L 309 Sc

R K H K S M I M A L P I L V L G F L A L K G G L R F T I Y S V P V M - - A L G F G F L L S E F - - - - - - - - - - - - - - - - - - - - - - - - - - - K A I L - - 348 Cj

N D E R V F I I L Y A V S A V Y F A G V M V R L M L T L T P A V C V L A G I G F S Y T F E K Y L K D E E T K E R S S S Q S G T T K D E K - - - - - - - - - L Y D 385 Ce

N D E R V F V V L Y A I S A V Y F A G V M V R L M L T L T P V V C M L A G V A F S G L L D V F L Q E D S S K R M G T A I S A A T E V D E A E D S I E K K T L Y D 384 Dm

S D A R I F I I M Y G V T S M Y F S A V M V R L M L V L A P V M C I L S G I G V S Q V L S T Y M K N L D I S R P D K - - - - - - - - - - - - - - - - - - - - - - 383 Hs

K D E H V F V I A Y S V L C S Y F A G V M V R L M L T L T P V I C V S A A V A L S K I F D I Y L - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 382 Sc

- - - - - - - - - V K K Y S Q L T S N V C I V F A T I L T L A P V - F I H I Y - - - - - - - - - - N Y K A P T V - - - F S Q N E A S - - - L L N Q - - - - - - - 397 Cj

K A A K N V K S R N A N D G D - E S G V S S N V R T I I S I I L V I F L L M F V V H A T Y V T S N A Y S H P S V V L Q - S S T N N G D R I I M D D F R E A Y H W 456 Ce

K A G K - L K H R T K H D A Q Q D T G V S S N L K S I V I L A V L M L L M M F A V H C T W V T S N A Y S S P S I V L A F H N S Q D G S R N I L D D F R E A Y Y W 464 Dm

- - - - - - - - K S K K Q Q D S T Y P I K N E V A S G M I L V M A F F L I T Y T F H S T W V T S E A Y S S P S I V L S - A R G G D G S R I I F D D F R E A Y Y W 441 Hs

- - - - D F K T S D R K Y A I K P A A L L A K L - - I V S G S F I F Y L Y L F V F H S T W V T R T A Y S S P S V V L P - S Q T P D G K L A L I D D F R E A Y Y W 430 Sc

L K N I A N R E D Y V V T W W D Y G Y P V R Y Y S D V K T L V D G G K H L G K D N F F P S F S L S K D E Q A A A N - M A R L S V E Y T E K S F Y A P Q N D I L K 444 Cj

L R E N T A D D A R V M S W W D Y G Y Q I A G M A N R T T L V D N N T W N N S H I A L V G K A M S S N E S A A Y E I M T E L D V D Y I L V I F - - - - - - - - - 534 Ce

L S Q N T A D D A R V M S W W D Y G Y Q I A G M A N R T T L V D N N T W N N S H I A L V G K A M S S T E E K S Y E I M T S L D V D Y V L V I F - - - - - - - - - 543 Dm

L R H N T P E D A K V M S W W D Y G Y Q I T A M A N R T I L V D N N T W N N T H I S R V G Q A M A S T E E K A Y E I M R E L D V S Y V L V I F - - - - - - - - - 512 Hs

L R M N S D E D S K V A A W W D Y G Y Q I G G M A D R T T L V D N N T W N N T H I A I V G K A M A S P E E K S Y E I L K E H D V D Y V L V I F - - - - - - - - - 503 Sc

S D I L Q A M M K D Y N Q S N V D L F L A S L S K P D F K I D T P K T R D I Y L Y M P A R M S L I F S T V A S F S F I N L D T G V L D K P F T F S T A Y P L D V 523 Cj

- - - - - G G V I G Y S G D D I N K F L W M V R I A Q - - - G E H P - K D I - - - - - - R E E N Y F T S T G E Y S T G A G A S E T M L N C L M Y K M S Y - - - Y 605 Ce

- - - - - G G V I G Y S G D D I N K F L W M V R I A E - - - G E H P - K D I - - - - - - K E S D Y F T D R G E F R V D A E G A P A L L N C L M Y K L S Y - - - Y 614 Dm

- - - - - G G L T G Y S S D D I N K F L W M V R I G G - - - S T D T G K H I - - - - - - K E N D Y Y T P T G E F R V D R E G S P V L L N C L M Y K M C Y - - - Y 583 Hs

- - - - - G G L I G F G G D D I N K F L W M I R I S E - - - G I W P E - E I - - - - - - K E R D F Y T A E G E Y R V D A R A S E T M R N S L L Y K M S Y - - - K 574 Sc

K N G E I Y L S - N G V V L S D D F R S F K I G D - N V V S V N S I V E I N S I K Q G E Y K I T P I D D K A Q F Y I F Y L K - - D S A I P Y A Q F I L M D K T M 603 Cj

R F G E T R V G Y N Q A G G F D R T R G Y V I G K K D I T - L E Y I E E A Y T T E N W L V R I Y K R K K L P - - N R P T V K S E E A T I P I K G - K K A T Q G K 667 Ce

R F G E L K L D Y R G P S G Y D R T R N A V I G N K D F D - L T Y L E E A Y T T E H W L V R I Y R V K K P H E F N R P S L K T K E R T I P P A N F I S R K N S K 676 Dm

R F G Q V Y T E A K R P P G F D R V R N A E I G N K D F E - L D V L E E A Y T T E H W L V R I Y K V K D L - - - - - - - - - - - - - - - - - - - - - - - - - - - 646 Hs

D F P Q L - - - F N G G Q A T D R V R Q Q M I T P L D V P P L D Y F D E V F T S E N W M V R I Y Q L K K D D - - - - - - - - A Q G R T L R D V G E L T R S S T K 636 Sc

F N S A Y V Q M F F L G N Y D K N L F D L V I N S R D A K V F K L K I 679 Cj

N K K G V I R - - - - - - - - - - - - P A P T A S K A 743 Ce

R R K G Y I R - - - - - - - - - - - - N R P V V V K G K R T L K 755 Dm

D N R G L S R - - - - - - - - - - - - T 698 Hs

T R R S I K R - - - - - - - - - - - - P E L G L R V 705 Sc

M L K K E Y L K - - - - - - N P Y L V L F A M I V L A Y V F - - S V F C R F Y W V W W - - - - A S E F N E Y F - F N N Q L M I I S N D G Y A F A E - - G A R D - 1 Cj

M T S T T A A R T A S S R V G A - T T L L T I V V L A L A W F V G F A S R L F A I V R F E S I I H E F D P W F N Y R A T H H M V Q H G F Y K F L N W F D E R A W 1 Ce

M N R T - - P K M L N S K V A G Y S S L I T F A I L L I A W L A G F S S R L F A V I R F E S I I H E F D P W F N Y R A T A Y M V Q N G W Y N F L N W F D E R A W 1 Dm

M T K F G F L R L S Y E K Q D - - - T L L K L L I L S M A A V L S F S T R L F A V L R F E S V I H E F D P Y F N Y R T T R F L A E E G F Y K F H N W F D D R A W 1 Hs

M G S D - - - - - R S C V L S V F Q T I L K L V I F V A I F G A A I S S R L F A V I K F E S I I H E F D P W F N Y R A T K Y L V N N S F Y K F L N W F D D R T W 1 Sc

- - - - - M I A G F H Q P N D L S Y Y G S S L S T L T Y W L Y K I T P F S F E S I I L Y M S T F L S S L V V I P I I L L A N E Y K R P L M G F V A A L L A S V A 65 Cj

Y P L G R I V G G T V Y P G L M V T S G L I H W I L D S - L - - N F H V H I R E V C V F L A P T F S G L T A I A T Y L L T K E L W S P G A G L F A A C F I A I S 80 Ce

Y P L G R I V G G T V Y P G L M I T S G G I H W L L H V - L - - N I P V H I R D I C V F L A P I F S G L T S I S T Y L L T K E L W S A G A G L F A A S F I A I V 79 Dm

Y P L G R I I G G T I Y P G L M I T S A A I Y H V L H F - F - - H I T I D I R N V C V F L A P L F S S F T T I V T Y H L T K E L K D A G A G L L A A A M I A V V 78 Hs

Y P L G R V T G G T L Y P G L M T T S A F I W H A L R N W L - - G L P I D I R N V C V L F A P L F S G V T A W A T Y E F T K E I K D A S A G L L A A G F I A I V 76 Sc

N S Y Y N R T M S G Y Y D T D M L V I V L P M F I L F F M V R M I L K K D F F - - S L I A L P L F I G I Y L W W Y P S S Y T L N V A L I G L F L I Y T L I F H R 140 Cj

P G Y T S R S V A G S Y D N E G I A I F A L Q F T Y Y L W V K S L K T G S I M W A S L C A L S Y F Y M V S A W - - - G G Y V F I I N L I P L H A L A L I I M G R 157 Ce

P G Y I S R S V A G S Y D N E G I A I F A L Q F T Y F L W V R S V K T G S V F W S A A A A L S Y F Y M V S A W - - - G G Y V F I I N L I P L H V F V L L I M G R 156 Dm

P G Y I S R S V A G S Y D N E G I A I F C M L L T Y Y M W I K A V K T G S I C W A A K C A L A Y F Y M V S S W - - - G G Y V F L I N L I P L H V L V L M L T G R 155 Hs

P G Y I S R S V A G S Y D N E A I A I T L L M V T F M F W I K A Q K T G S I M H A T C A A L F Y F Y M V S A W - - - G G Y V F I T N L I P L H V F L L I L M G R 154 Sc

- K E K I F - - - - - - Y I A V I L S S L T L S N I A W F - Y Q S A I I V I L F A L F A L E Q - - - - - - - - - - - - K R L N F M I I - - - - - - - - - - G I L 218 Cj

Y S S R L F V S Y T S F Y C L A T I L S M Q V P F V G F Q P V R T S E H M P A F G V F G L L Q I V A L M H Y A R N R I T R Q Q F M T L F V - G G L T I - L G A L 234 Ce

Y S P R L L T S Y S T F Y I L G L L F S M Q I P F V G F Q P I R T S E H M A A L G V F V L L M A V A T L R H L Q S V L S R N E F R K L F I V G G L L V G V G V F 233 Dm

F S H R I Y V A Y C T V Y C L G T I L S M Q I S F V G F Q P V L S S E H M A A F G V F G L C Q I H A F V D Y L R S K L N P Q Q F E V L F R S V I S L V G F V L L 232 Hs

Y S S K L Y S A Y T T W Y A I G T V A S M Q I P F V G F L P I R S N D H M A A L G V F G L I Q I V A F G D F V K G Q I S T A K F K V I M M V S L F L I - - L V L 231 Sc

G S A T L I F L I L S G G V D P I L Y Q L K F Y I F R S D E S A N L T Q G F M Y F N V N Q T I Q E V E N V D F S E F M R R I S G S E I V F L F S L F G F V W L L 268 Cj

S V V V Y F A L V W G G Y V A P - - F S G R F Y S L W D T G Y A K - - - - - I H I P I I A S V S E H Q P T T W V S F F F D L H I T A A V F P V G L W Y C I K K V 312 Ce

V A V V V L T M L - - G V V A P - - W S G R F Y S L W D T G Y A K - - - - - I H I P I I A S V S E H Q P T T W F S F F F D L H I L V C A F P V G V W Y C I K Q I 313 Dm

T V G A L L M L - - T G K I S P - - W T G R F Y S L L D P S Y A K - - - - - N N I P I I A S V S E H Q P T T W S S Y Y F D L Q L L V F M F P V G L Y Y C F S N L 312 Hs

G V V G L S A L T Y M G L I A P - - W T G R F Y S L W D T N Y A K - - - - - I H I P I I A S V S E H Q P V S W P A F F F D T H F L I W L F P A G V F L L F L D L 309 Sc

R K H K S M I M A L P I L V L G F L A L K G G L R F T I Y S V P V M - - A L G F G F L L S E F - - - - - - - - - - - - - - - - - - - - - - - - - - - K A I L - - 348 Cj

N D E R V F I I L Y A V S A V Y F A G V M V R L M L T L T P A V C V L A G I G F S Y T F E K Y L K D E E T K E R S S S Q S G T T K D E K - - - - - - - - - L Y D 385 Ce

N D E R V F V V L Y A I S A V Y F A G V M V R L M L T L T P V V C M L A G V A F S G L L D V F L Q E D S S K R M G T A I S A A T E V D E A E D S I E K K T L Y D 384 Dm

S D A R I F I I M Y G V T S M Y F S A V M V R L M L V L A P V M C I L S G I G V S Q V L S T Y M K N L D I S R P D K - - - - - - - - - - - - - - - - - - - - - - 383 Hs

K D E H V F V I A Y S V L C S Y F A G V M V R L M L T L T P V I C V S A A V A L S K I F D I Y L - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 382 Sc

- - - - - - - - - V K K Y S Q L T S N V C I V F A T I L T L A P V - F I H I Y - - - - - - - - - - N Y K A P T V - - - F S Q N E A S - - - L L N Q - - - - - - - 397 Cj

K A A K N V K S R N A N D G D - E S G V S S N V R T I I S I I L V I F L L M F V V H A T Y V T S N A Y S H P S V V L Q - S S T N N G D R I I M D D F R E A Y H W 456 Ce

K A G K - L K H R T K H D A Q Q D T G V S S N L K S I V I L A V L M L L M M F A V H C T W V T S N A Y S S P S I V L A F H N S Q D G S R N I L D D F R E A Y Y W 464 Dm

- - - - - - - - K S K K Q Q D S T Y P I K N E V A S G M I L V M A F F L I T Y T F H S T W V T S E A Y S S P S I V L S - A R G G D G S R I I F D D F R E A Y Y W 441 Hs

- - - - D F K T S D R K Y A I K P A A L L A K L - - I V S G S F I F Y L Y L F V F H S T W V T R T A Y S S P S V V L P - S Q T P D G K L A L I D D F R E A Y Y W 430 Sc

L K N I A N R E D Y V V T W W D Y G Y P V R Y Y S D V K T L V D G G K H L G K D N F F P S F S L S K D E Q A A A N - M A R L S V E Y T E K S F Y A P Q N D I L K 444 Cj

L R E N T A D D A R V M S W W D Y G Y Q I A G M A N R T T L V D N N T W N N S H I A L V G K A M S S N E S A A Y E I M T E L D V D Y I L V I F - - - - - - - - - 534 Ce

L S Q N T A D D A R V M S W W D Y G Y Q I A G M A N R T T L V D N N T W N N S H I A L V G K A M S S T E E K S Y E I M T S L D V D Y V L V I F - - - - - - - - - 543 Dm

L R H N T P E D A K V M S W W D Y G Y Q I T A M A N R T I L V D N N T W N N T H I S R V G Q A M A S T E E K A Y E I M R E L D V S Y V L V I F - - - - - - - - - 512 Hs

L R M N S D E D S K V A A W W D Y G Y Q I G G M A D R T T L V D N N T W N N T H I A I V G K A M A S P E E K S Y E I L K E H D V D Y V L V I F - - - - - - - - - 503 Sc

S D I L Q A M M K D Y N Q S N V D L F L A S L S K P D F K I D T P K T R D I Y L Y M P A R M S L I F S T V A S F S F I N L D T G V L D K P F T F S T A Y P L D V 523 Cj

- - - - - G G V I G Y S G D D I N K F L W M V R I A Q - - - G E H P - K D I - - - - - - R E E N Y F T S T G E Y S T G A G A S E T M L N C L M Y K M S Y - - - Y 605 Ce

- - - - - G G V I G Y S G D D I N K F L W M V R I A E - - - G E H P - K D I - - - - - - K E S D Y F T D R G E F R V D A E G A P A L L N C L M Y K L S Y - - - Y 614 Dm

- - - - - G G L T G Y S S D D I N K F L W M V R I G G - - - S T D T G K H I - - - - - - K E N D Y Y T P T G E F R V D R E G S P V L L N C L M Y K M C Y - - - Y 583 Hs

- - - - - G G L I G F G G D D I N K F L W M I R I S E - - - G I W P E - E I - - - - - - K E R D F Y T A E G E Y R V D A R A S E T M R N S L L Y K M S Y - - - K 574 Sc

K N G E I Y L S - N G V V L S D D F R S F K I G D - N V V S V N S I V E I N S I K Q G E Y K I T P I D D K A Q F Y I F Y L K - - D S A I P Y A Q F I L M D K T M 603 Cj

R F G E T R V G Y N Q A G G F D R T R G Y V I G K K D I T - L E Y I E E A Y T T E N W L V R I Y K R K K L P - - N R P T V K S E E A T I P I K G - K K A T Q G K 667 Ce

R F G E L K L D Y R G P S G Y D R T R N A V I G N K D F D - L T Y L E E A Y T T E H W L V R I Y R V K K P H E F N R P S L K T K E R T I P P A N F I S R K N S K 676 Dm

R F G Q V Y T E A K R P P G F D R V R N A E I G N K D F E - L D V L E E A Y T T E H W L V R I Y K V K D L - - - - - - - - - - - - - - - - - - - - - - - - - - - 646 Hs

D F P Q L - - - F N G G Q A T D R V R Q Q M I T P L D V P P L D Y F D E V F T S E N W M V R I Y Q L K K D D - - - - - - - - A Q G R T L R D V G E L T R S S T K 636 Sc

F N S A Y V Q M F F L G N Y D K N L F D L V I N S R D A K V F K L K I 679 Cj

N K K G V I R - - - - - - - - - - - - P A P T A S K A 743 Ce

R R K G Y I R - - - - - - - - - - - - N R P V V V K G K R T L K 755 Dm

D N R G L S R - - - - - - - - - - - - T 698 Hs

T R R S I K R - - - - - - - - - - - - P E L G L R V 705 Sc

A

B

Figure 2.3: Hydropathy blot and alignment of the amino acid sequences of Stt3 homologs. (A)Hydropathy alignment of the amino acid sequences of Stt3p and PglB. The membrane topology ishighly conserved. The proteins have a size of about 80 kDa, containing a cleavable signal sequence, 10-12 predicted transmembrane segments and a hydrophilic C-terminus covering about one third of theprotein. (B) An alignment of the amino acid sequences of PglB and its homologs. The sequences are:Cj, PglB of Campylobacter jejuni (GenBank accession number NP282274); Ce, T12A2.2p ofCaenorhabditis elegans (accession number NP498362); Hs, Stt3p homolog of Homo sapiens(accession number XP006242); Dm, OstStt3p of Drosophila melanogaster (accession numberNP524494); Sc, Stt3p of Saccharomyces cerevisiae (accession number NP011493). The highlyconserved WWDYGY motif is boxed.

37

2.1.6 The pgl locus is essential for N-linked protein glycosylation

in C. jejuni: a hypothesis

It has been noticed previously (Bugg and Brandish 1994) that the LPS biosynthesis in

bacteria and the N-linked protein glycosylation pathway in the ER of eukaryotes share

significant similarities. Both pathways initiate with the assembly of an

oligosaccharide on a carrier lipid, bactoprenol in the case of the bacterial pathway

(Raetz 1996; Whitfield 1995; Whitfield and Roberts 1999) and dolichol for the

eukaryotic biosynthesis (Burda et al., 1999). The oligosaccharide assembly takes

place at the cytoplasmic side of the membrane and nucleotide-activated sugars serve

as donor substrates for specific glycosyltransferases. These transferases can share

significant sequence identity between the different domains. In both pathways, the

lipid-linked oligosaccharide is translocated across the membrane. In the ER of

eukaryotic cells, assembly of the oligosaccharide continues before it is transferred to

selected acceptor sites on proteins, whereas in bacteria, lipid-linked oligosaccharides

serve as building-blocks for LPS biosynthesis. Based on these similarities, a

hypothetical model for protein glycosylation in C. jejuni is proposed (Figure 2.4): it is

suggested that C. jejuni is able to synthesize N-glycoproteins and that the pgl locus is

responsible for the synthesis of protein-linked oligosaccharides.

In the first step of this putative glycosylation pathway, PglC attaches a

monosaccharide-P to undecaprenyl phosphate, using a nucleotide sugar precursor.

PglC shows homology to proteins that are involved in the synthesis of the antigen unit

of LPS molecules. They transfer the nucleotide activated sugar to the carrier lipid.

Hydrophobicity plots of these proteins suggest a similar pattern of secondary

structures consisting of several transmembrane helices (Bugg and Brandish 1994).

The highest similarity to PglC was found for WbaP of Salmonella enterica that

transfers Gal-P to undecaprenyl phosphate (Fry et al., 1998). The other proteins that

are encoded by genes of the pgl locus and show homologies to glycosyltransferases

are responsible for the assembly of full length lipid-linked oligosaccharide (LLO).

WlaB, which shows homologies to ABC transporters (Fry et al., 1998), is proposed to

translocate the LLO into the periplasmic space. One of the WlaB-homologs, MsbA of

E. coli, is an essential protein and is involved in the transport of nascent core-lipid A

molecules across the inner membrane (Zhou et al., 1998).

38

The model postulates that the putative OTase PglB (the Stt3p homolog) is responsible

for glycosylation of proteins with the consensus sequences Asn-X-Ser/Thr. It transfers

the oligosaccharide from the carrier lipid to the protein in the essential reaction of the

process. This reaction takes place in the periplasm, the equivalent to the lumen of the

ER of eukaryotes.

PglB

PglC PglXY

Cytoplasm

Periplasm

PP

PP

Asn

PPPWlaB

galE wlaB pglH pglI pglJ pglB pglA pglC pglD pglE pglF pglG

PglB

PglC PglXY

Cytoplasm

Periplasm

PP

PP

Asn

PPPWlaB

galE wlaB pglH pglI pglJ pglB pglA pglC pglD pglE pglF pglG

Figure 2.4: Proposed model for N-linked protein glycosylation pathway in C. jejuni. As ineukaryotes an oligosaccharide is assembled on a carrier lipid. First, PglC attaches the first sugar-P topolyprenyl phosphate, the glycosyltransferase homologs encoded by genes of the pgl locus thencomplete the synthesis of the LLO, and the LLO is flipped into the periplasmic space, a processcatalyzed by the ABC transporter homolog WlaB. In the central reaction of the process, PglB, theOTase, catalyzes the formation of the sugar-protein bond. This final reaction takes place in theperiplasm, the equivalent to the lumen of the ER in eukaryotes.

39

2.2 Results

2.2.1 Chemical deglycosylation affects protein antigenicity

It was shown previously that the pgl locus in C. jejuni is involved in protein

glycosylation (Linton et al., 2002; Szymanski et al., 1999). In particular, mutagenesis

of pgl genes results in a dramatic reduction of immunoreactivity of multiple C. jejuni

proteins with an antiserum raised against C. jejuni whole cell extracts. A similar loss

of immunoreactivity is evident following treatment of C. jejuni proteins with

trifluoromethanesulphonic acid (TFMS), a deglycosylating agent (Szymanski et al.,

1999). It was concluded that the glycosyl moieties of the C. jejuni proteins are

immunodominant. To test if the polyclonal antiserum that was raised against C. jejuni

whole cell extracts and was used during the studies of this thesis also recognizes

glycoproteins, C. jejuni membrane proteins were treated with TFMS. TFMS

chemically deglycosylates glycoproteins with varying efficiency, depending on the

sugar linkage and the type of sugar involved, while leaving the protein backbone

intact.

After deglycosylation of membrane proteins with TFMS, the samples were separated

by SDS-PAGE and were either stained by silver (Figure 2.5, lanes 3 and 4) or, after

blotting to nitrocellulose membrane, incubated with the polyclonal antiserum (lanes 1

and 2). It is apparent that after treatment with TFMS, immunoreactivity of specific

proteins disappeared (lane 2 compared to untreated extracts, lane 1), whereas the

overall protein pattern as visualized by silver staining was not affected by the

chemical treatment (lane 4 compared to untreated extracts, lane 3). These results

confirmed that C. jejuni expressed glycoproteins in low amounts and that the

polyclonal antiserum specifically recognized these glycoproteins.

40

+- +-TFMS

4321 4321

47.5

32.5

175

83

62

25

kDa

47.5

32.5

175

83

62

25

kDa

47.5

32.5

175

83

62

25

kDa

Figure 2.5: Chemical deglycosylation of membrane proteins of C. jejuni using TFMS. Untreatedand treated membrane proteins were separated by SDS-PAGE and either after blotting immunodetectedwith the C. jejuni antiserum (lane 1 + 2) or stained by silver (lane 3 + 4). Lane 1 + 3, untreatedmembrane proteins; lane 2 + 4 TFMS treated membrane proteins. Molecular weight markers areindicated on the left.

2.2.2 Hydrophilic C-terminal domain of PglB is located in the

periplasm

Within the frameworks of the hypothesis that the Stt3p homolog PglB acts as an

OTase in the N-linked protein glycosylation pathway of C. jejuni, it is proposed that

the topology of PglB is similar to the one observed for Stt3p. Indeed, both proteins

carry a putative N-terminal signal sequence, locating the N-terminus of the mature

protein in the lumen of the ER/periplasmic space. It was shown for the yeast Stt3p

that the hydrophilic C-terminal domain is located in the lumen of the ER (Zufferey et

al., 1995). It was therefore tested if the C-terminus of C. jejuni PglB was placed in the

periplasm. To determine the topology of PglB at the plasma membrane, the reporter

molecule alkaline phosphatase A (PhoA) of E. coli was attached to hydrophilic

domains of the membrane protein and expression of the resulting chimeric proteins

checked in E. coli. PhoA is a metalloenzyme that consists of two identical subunits

and resides in the periplasm (Chang et al., 1986). Each of the subunits contains two

intramolecular disulfide bridges that trigger the correct folding of PhoA. Therefore,

41

the process of folding and assembly of PhoA occurs only after export to the

periplasmic space (Akiyama and Ito 1993). This periplasm specific character of PhoA

allows its use as a reporter molecule (van Geest and Lolkema 2000) and to study the

membrane topology of PglB when expressed in E. coli.

The mature PhoA protein, lacking its signal sequence, was fused to 5 different C-

terminally truncated forms of the membrane protein PglB. If PhoA was fused to a

domain that is normally located in the cytoplasm, the PhoA molecule also remained in

the cytoplasm and was therefore inactive. In contrast, when PhoA was fused to a

domain that is located in the periplasm, the PhoA moiety of the fusion protein was

also exported across the membrane and was folded and assembled into the active

state. Plasmids expressing the fusion proteins were transformed into E. coli host cells

that contained a deletion in the genomic copy of phoA and therefore did not produce

any endogenous PhoA protein (data not shown). Protein extracts of these cells were

separated by SDS-PAGE and transferred to nitrocellulose membrane. To check the

expression of the fusion proteins, the Western blot was incubated with an antiserum

against PhoA (Figure 2.6A). Cells transformed with the five constructs expressed a

chimeric protein, as revealed by reactivity against the antiserum. The fusion proteins

showed an aberrant mobility in SDS-PAGE (Figure 1A, lane 1 (predicted size 54.4

kDa), lane 2 (82.5 kDa), lane 3 (89.6 kDa), lane 4 (98.8 kDa), lane 5 (128.9 kDa)).

However, hydrophobic membrane proteins are known for their abnormal mobility in

SDS-PAGE. Furthermore, fusion proteins were degraded. The arrow in Figure 2.6A

points to a protein that runs with a mobility of around 48 kDa in SDS-PAGE and is

recognized by the antiserum. This correlates with the expected size of PhoA.

It was shown that the five different fusion proteins were expressed in E. coli. As

mentioned before, the enzymatic activity of the fusion protein reveals the cellular

location of the fusion site. Therefore, PhoA activity of cells expressing the fusion

constructs was measured in vitro using p-nitrophenyl-phosphate (pNPP) as a substrate

(Figure 2.6A). All five different fusion proteins showed activity in vitro and the

activity correlated with the expression levels. The smallest protein (lane 1, 58.5 kDa)

was expressed at the highest level and showed the highest in vitro PhoA activity,

probably because it was predicted to be a soluble periplasmic protein. The other four

fusion proteins that were predicted to contain various membrane domains were

expressed in lower amounts and showed lower in vitro activities. Control cells that did

not express PhoA showed only background activity (Figure 2.6A, lane 6). Based on

42

these results, the following topology model for PglB was proposed (Figure 2.6B): the

polypeptide consists of 10 transmembrane domains and a hydrophilic C-terminus. The

C-terminus is located in the periplasm and contains the most highly conserved amino

acid motif (see also Figure 2.7A).

43

Cytoplasm

Periplasm

A

B

�PhoA

654321

47.5

175

83

62

kDa

�PhoA

654321

47.5

175

83

62

kDa

654321 654321

47.5

175

83

62

kDa

47.5

175

83

62

kDa

1 2

3

4 5

54.4

kD

a

82.5

kD

a

89.6

kD

a

98.8

kD

a

128.

9 kD

a

299 22 47 38 34 11

in vitro PhoA activity (Miller Units)

Figure 2.6: Membrane topology of PglB-PhoA fusion constructs in E. coli. (A) The fusionconstructs (see also B) had a molecular weight (in kDa) of 54.5 (1), 82.5 (2), 89.6 (3), 98.8 (4), and128.9 (5). In vitro PhoA activity of cells expressing the fusion proteins was determined with p-nitrophenyl-phosphate at OD420. The activity is shown in Miller units. Cells expressing the fusionproteins were applied to the SDS-PAGE in the same order. Whole cell extracts of E. coli CC118(∆phoA) carrying the plasmids for the expression of the fusion proteins were separated by SDS-PAGEand transferred to a nitrocellulose membrane. The expression of the fusion proteins was probed with anantiserum against PhoA. The arrow shows the mobility of the degradation product representing thecomplete PhoA domain. Molecular weight markers are indicated on the left. Based on the expressionand in vitro activity, a topological model is proposed (B) where the N-terminus and the highlyconserved C-terminal domain of PglB are located in the periplasm. Arrows show the locations of thefusion constructs.

44

2.2.3 PglB plays a major role in protein glycosylation in C. jejuni

To address directly the function of PglB in protein glycosylation, a deletion of the

corresponding gene in C. jejuni was made by replacing most of its coding sequence

with a kanamycin resistance (Kmr) cassette. This cassette contains the aphA gene of

Campylobacter coli (Trieu-Cuot et al., 1985) and the product of the aphA gene

confers resistance to kanamycin. The aphA gene contains its own promoter but lacks

transcription termination signals. To ensure expression of the pgl genes downstream

of pglB, the aphA gene was inserted in an orientation such that the genes downstream

of the mutated pglB would be transcribed by the aphA promoter. The correct insertion

in the mutant strain was confirmed by PCR (data not shown). When membrane

proteins (Figure 2.7B) of WT (lane 1) and pglB mutant (lane 2) cells were separated

by SDS-PAGE and visualized by silver staining, no difference in the protein pattern

was detected. In contrast, after transfer to a nitrocellulose membrane and incubating

with the polyclonal antiserum that specifically recognizes glycoproteins (Figure 2.5),

a loss of affinity to some membrane proteins of pglB mutant cells was evident (Figure

2.7C, lane 2) as compared to WT proteins (lane 1). The deletion of pglB led to a

change in the recognition of proteins by the antiserum. Two immunodominant

proteins (Figure 2.7C, lane 1) with mobility in SDS-PAGE of 47 kDa and 35 kDa

were the most prominent proteins to be affected in the mutant strain. They are marked

by arrows in Figure 2.7C.

To ascertain that the deletion of the pglB coding sequence was responsible for the

observed phenotype, the pglB mutation was complemented with a plasmid expressing

PglB. The 16 kbp SalI-NcoI fragment containing most of the pgl locus from pBTLPS

(Fry et al., 1998) was cloned into pRY111, a chloramphenicol-resistant E. coli-

Campylobacter shuttle plasmid (Yao et al., 1993), to generate pRY111(pgl). This

plasmid was conjugated from E. coli DH5α (RK212.1) into pgl mutant cells. The

electrophoretic profile of membrane proteins of a resulting strain is shown in Figure

2.7B/C (lane 3). Immunoreactivity of proteins absent in the pglB mutant was partially

restored by the introduction of pglB in trans (Figure 2.7C, lane 3).

It was postulated that the highly conserved WWDYGY motif of PglB (Figure 2.7A)

plays an essential role in glycosylation. To test the effect of point mutations in this

motif on glycosylation, plasmids expressing PglB with amino acid exchanges in this

region were conjugated into the pglB mutant strain (Figure 2.7B/C lane 4-7). The

45

amino acids of the highly conserved motif that were exchanged are marked with a

point in Figure 2.7A. Four different plasmids were constructed that express PglB with

the following amino acid exchanges: W458A, D459A; Y460E, G461E; Y460E;

G461E. Silver staining of total membrane proteins of these strains after SDS-PAGE

did not show a difference as compared to proteins of WT or pglB mutant cells (Figure

2.7B, lane 4-7). Most interestingly, when membrane proteins of mutant cells

expressing PglB with amino acid exchanges were separated by SDS-PAGE and

transferred to nitrocellulose, the antiserum specific for glycoproteins did not detect the

specific proteins as compared to proteins of WT and of mutant cells that express PglB

(Figure 2.7C). When Trp and Asp or Tyr and Gly were exchanged in PglB (Figure

2.7C, lane 4 (W458A, D459A) and lane 5 (Y460E, G461E)), PglB did not

complement pglB phenotype (Figure 2.7C, lane 2). In contrast, exchange of either Tyr

or Gly (Y460E or G461E) could partially complement the deficiency of the mutant.

Lane 6 shows that the exchange of Tyr to Glu led to the presence of two

immunoreactive proteins at around 47 kDa in contrast to the pglB mutant where both

of these proteins were not detected. The exchange of the Gly to Glu (Figure 2.7C, lane

7) led to the presence of only the faster migrating band at 47 kDa. It was postulated

that the slowest migrating band represented a protein containing two covalently linked

oligosaccharides, whereas the faster migrating band was the monoglycosylated form.

In WT cells, the diglycosylated protein was produced, whereas amino acid exchanges

in the highly conserved C-terminal domain of PglB led to production of

monoglycosylated and unglycosylated forms.

46

B

47.5

32.5

16.5

175

8362

25

kDa

47.5

32.5

16.5

175

8362

25

kDa

47.5

32.5

16.5

175

8362

25

kDa

pglB-

p(pgl)

W4 5

8A, D

459A

Y46

0E, G

461 E

Y46

0 E

G46

1 E

WT

47.5

32.5

16.5

17583

62

25

C

7654321

47.5

32.5

16.5

17583

62

25

47.5

32.5

16.5

17583

62

25

C

7654321

C

7654321 7654321 7654321

A 544 V M S W W D Y G Y Q I A Ce 522 V M S W W D Y G Y Q I T Hs 553 V M S W W D Y G Y Q I A Dm 513 V A A W W D Y G Y Q I G Sc 454 V V T W W D Y G Y P V R Cj

544 V M S W W D Y G Y Q I A Ce 522 V M S W W D Y G Y Q I T Hs 553 V M S W W D Y G Y Q I A Dm 513 V A A W W D Y G Y Q I G Sc 454 V V T W W D Y G Y P V R Cj

Figure 2.7: Comparison of membrane proteins of C. jejuni WT and pglB mutant andcomplementation in trans. (A) An alignment of the highly conserved amino acid motif of PglB and itshomologs. The sequences are: Ce, T12A2.2p of Caenorhabditis elegans (GenBank accession numberNP498362); Hs, Stt3p homolog of Homo sapiens (accession number XP006242); Dm, OstStt3p ofDrosophila melanogaster (accession number NP524494); Sc, Stt3p of Saccharomyces cerevisiae(accession number NP011493); Cj, PglB of Campylobacter jejuni (accession number NP282274). Thepoints mark the amino acids that are mutated in the different constructs. Membrane proteins wereseparated by SDS-PAGE and either stained by silver (B) or immunodetected with the C. jejuniantiserum (C). Lane 1, WT; lane 2, pglB mutant; lane 3, pglB mutant containing a plasmid thatexpresses PglB; lane 4-7, pglB mutant containing a plasmid that expresses PglB with amino acidexchanges in the highly conserved hydrophilic C-terminal domain. The following amino acids wereexchanged: lane 4, W458A and D459A; lane 5, Y460E and G461E; lane 6, Y460E; lane 7, G461E.Molecular weight markers are indicated on the left. The arrows show the positioning of the mostimmunogenic putative glycoproteins, which were purified.

47

2.2.4 The locus with pglB mutations complements the pglE

deletion

To examine the influence of other proteins on protein glycosylation that are encoded

by the genes of the pgl locus, most of the open reading frame of pglE was deleted by

introduction of the Kmr cassette through homologous recombination (Szymanski et

al., 1999). PglE is homologous to aminotransferases involved in LPS biosynthesis. It

was assumed that this enzyme is required for the synthesis of a nucleotide sugar that

serves as a precursor in the synthesis of the lipid-linked oligosaccharide.

Membrane proteins of WT and pglE mutant cells were separated by SDS-PAGE

(Figure 2.8). There was no difference detected when the proteins were visualized by

silver staining (Figure 2.8A, lane 1 + 2). However, after transfer of the proteins to

nitrocellulose membrane and incubation with the antiserum specific for glycoproteins,

altered glycoprotein expression was observed in pglE mutant cells (Figure 2.8B, lane

2 as compared to lane 1). Though the same set of immunoreactive glycoproteins was

affected in the pglE mutant as in the pglB mutant, the severity of phenotype was less

pronounced. The mutant pattern could be complemented by introduction of the

plasmid containing the pgl locus. Plasmids expressing WT PglB and expressing PglB

with amino acid exchanges in the highly conserved region could complement the pglE

mutant to the same extent (Figure 2.8B, lane 3 - 7). When membrane proteins of pglF-

deleted cells (Szymanski et al., 1999) were separated by SDS-PAGE and incubated

with the antiserum specific for glycoproteins, a similar loss of affinity as in pglE

mutation was observed (data not shown).

These results showed that two other proteins besides PglB encoded by pgl genes were

involved in protein glycosylation in C. jejuni. Based on homologies, it was proposed

that PglE and PglF are involved in the synthesis of a sugar nucleotide, which could

serve as precursor for the synthesis of lipid-linked oligosaccharide. This would

explain the underglycosylation of proteins observed in cells with a pglE or pglF

mutation. However, there was not a complete loss of glycosylation as in the pglB

mutant. A similar observation has been made in the eukaryotic system: mutations

affecting the biosynthesis of the lipid-linked oligosaccharide lead to a partial

glycosylation (due to relaxed specificity of the OTase (Aebi et al., 1996; Burda et al.,

1996; Burda and Aebi 1998; Burda et al., 1999; Jackson et al., 1993; Reiss et al.,

48

1996; Stagljar et al., 1994; te Heesen et al., 1994)), whereas inactivation of the OTase

results in complete loss of glycoprotein synthesis.

pglE-

p(pgl)

W45

8A, D

4 59 A

Y4 6

0E, G

461 E

Y46

0 E

G4 6

1E

WT

pglE-

p(pgl)

W45

8A, D

4 59 A

Y4 6

0E, G

461 E

Y46

0 E

G4 6

1E

WT

p(pgl)

W45

8A, D

4 59 A

Y4 6

0E, G

461 E

Y46

0 E

G4 6

1E

WT

A

B

7654321 7654321

47.5

32.5

16.5

17583

62

25

47.5

32.5

16.5

17583

62

25

47.5

32.5

16.5

17583

62

25

47.5

32.5

16.5

17583

62

25

47.5

32.5

16.5

17583

62

25

47.5

32.5

16.5

17583

62

25

47.5

32.5

16.5

1758362

25

kDa

47.5

32.5

16.5

1758362

25

kDa

47.5

32.5

16.5

1758362

25

kDa

Figure 2.8: Comparison of membrane proteins of C. jejuni WT and pglE mutant andcomplementation in trans. Membrane proteins were separated by SDS-PAGE and either stained bysilver (A) or immunodetected with the C. jejuni antiserum (B). Lane 1, WT; lane 2, pglE mutant; lane3, pglE mutant with a plasmid containing the pgl locus; lane 4-7, pglE mutant containing a plasmid thatexpresses PglB with the following amino acid exchanges in the highly conserved hydrophilic C-terminal domain: lane 4, W458A and D459A; lane 5, Y460E and G461E; lane 6, Y460E; lane 7,G461E. Molecular weight markers are indicated on the left.

49

2.2.5 Purification and identification of two glycoproteins of C.

jejuni

To identify putative glycoproteins of C. jejuni, the immunoreactive proteins of

approximate size of 47 kDa and 35 kDa (arrows in Figure 2.7C and 2.9A) were

chosen for purification. Membrane proteins of C. jejuni were solubilized and purified

by DEAE and soybean agglutinin (SBA) affinity columns as outlined in “Material and

Methods”. In Figure 2.9A, the purification was analyzed by separation of the proteins

by SDS-PAGE and visualization by silver staining, whereas in Figure 2.9B, proteins

were transferred to nitrocellulose and incubated with the polyclonal antiserum that

recognizes glycoproteins. Membrane proteins of C. jejuni were isolated (Figure

2.9A/B, lane 3) and solubilized in 0.4% n-dodecyl-β-D-maltoside (lane 4).

Solubilized membrane proteins were separated by anionic exchange column using a

gradient ranging from 10 mM NaCl to 1 M NaCl. Immunogenic proteins were eluted

from the column with 10 mM NaCl (lane 5) and applied to a SBA agglutinin agarose

column. SBA agglutinin is a lectin that recognizes glycoproteins from C. jejuni

(Linton et al., 2002). Proteins that did not show affinity for the lectin were also not

recognized by the antiserum that is specific for glycoproteins (lane 6). Proteins that

were bound by the lectin were eluted with 0.5 M galactose (lane 7) and these proteins

were recognized by the antiserum. This finding supported the hypothesis that the

antiserum raised against Campylobacter whole cell extracts was specific for

glycoproteins (see figure 2.5).

50

A

47.5

32.5

175

83

62

25

kDa

47.5

32.5

175

83

62

25

kDa

47.5

32.5

175

83

62

25

kDa

B

7654321

47.5

32.5

175

83

62

25

7654321 7654321

47.5

32.5

175

83

62

25

47.5

32.5

175

83

62

25

Figure 2.9: Purification of two putative glycoproteins of C. jejuni. To follow purification,membrane proteins were separated by SDS-PAGE and either stained by silver (A) or immunodetectedwith the C. jejuni antiserum (B). Lane 1, WT; lane 2, pglB mutant; lane 3, WT; lane 4, membraneproteins solubilized in 0.4% n-Dodecyl-β-D-maltoside; lane 5, fraction that did not bind to DEAEsepharose column, lane 6, fraction that did not bind to SBA column; lane 7, glycoproteins eluted fromSBA column with 0.5 M galactose. Molecular weight markers are indicated on the left. The arrowsshow the position of the most immunogenic putative glycoproteins that were cut from the gel andidentified.

Eluted glycoproteins from the lectin column were separated by SDS-PAGE and

stained by Coomassie. The two immunogenic proteins of an approximate size of 47

kDa and 35 kDa (marked in Figure 2.9A) were cut from the gel and digested with

trypsin. Peptide fragments were subjected to matrix-assisted laser

desorption/ionization (MALDI) mass spectrometry. Proteins were identified by

peptide mass fingerprinting and database searching, using the TrEMBL-database on

ExPasy. The protein migrating at 47 kDa was identified as a putative periplasmic

membrane fusion component of an efflux system, annotated as Cj0367c in the C.

jejuni NCTC 11168 genome sequence. The gene product shows homology to the E.

51

coli AcrA protein and therefore this name was proposed for the C. jejuni protein. The

second protein, migrating at 35 kDa, was identified as the OmpH1 protein

(Meinersmann et al., 1997), annotated as Cj0982c in the C. jejuni NCTC 11168

genome sequence. Table II gives the details of the peptide-mass fingerprinting

analysis.

Mr of

band

(kDa)

Expasy

identification

Percentage

of mass-

matched

peptides

Location and sequences of mass-matched

peptides (indicated in bold)

47 AcrA 62% MKLFQKNTILALGVVLLLAACSKEEAPKIQMPPQPVTTMSAKSEDLPLSFTYPAKLVSDYDVIIKPQVSGVIVNKLFKAGDKVKKGQTLFIIEQDKFKASVDSAYGQALMAKATFENASKDFNRSKALFSKSAISQKEYDSSLATFNNSKASLASARAQLANARIDLDHTEIKAPFDGTIGDALVNIGDYVSASTTELVRVTNLNPIYADFFISDTDKLNLVRNTQSGKWDLDSIHANLNLNGETVQGKLYFIDSVIDANSGTVKAKAVFDNNNSTLLPGAFATITSEGFIQKNGFKVPQIGVKQDQNDVYVLLVKNGKVEKSSVHISYQNNEYAIIDKGLQNGDKIILDNFKKIQVGSEVKEIGAQ

35 OmpH1 46% MKKILLSVLTTFVAVVLAACGGNSDSKTLNSLDKIKQNGVVRIGVFGDKPPFGYVDEKGNNQGYDIALAKRIAKELFGDENKVQFVLVEAANRVEFLKSNKVDIILANFTQTPERAEQVDFCLPYMKVALGVAVPKDSNITSVEDLIDKTLLLNKGTTADAYFTQDYPNIKTLKYDQNTETFAALMDKRGDALSHDNTLLFAWVKDHPDFKMGIKELGNKDVIAPAVKKGDKELKEFIDNLIIKLGQEQFFHKAYDETLKAHFGDDVKADDVVIEGGKI

Table II: Summary of peptide-mass fingerprinting data. Potential N-glycosylation sites areunderlined.

AcrA indeed contains 4 consensus sequences for putative N-linked protein

glycosylation. The sites with the sequence Asn-X-Ser/Thr are underlined in Table II.

One peptide containing a consensus sequence was found to be unmodified based on

the results from the mass spectrometry. The peptides containing the other 3

glycosylation sites were not found by mass spectrometry. These results supported the

hypothesis that C. jejuni contains N-linked glycoproteins. Based on previous results

(Figure 2.7C and 2.8B), it was postulated that two of the three remaining consensus

sequences of AcrA are modified in vivo.

52

AcrA of C. jejuni 81-176 was sequenced and the gene product compared to the protein

sequence of AcrA of C. jejuni NCTC 11168 strain (Parkhill et al., 2000) (Figure

2.10). There are 6 amino acid exchanges, the AcrA of C. jejuni 81-176 strain has a

calculated relative molecular weight of 40,004 Da and an isoelectric point (pI) of

8.29. Database similarity searches with the amino acid sequence of the mature protein

identified significant levels of similarity to membrane fusion proteins of different

efflux systems. It shows 31% identity to AcrA of E. coli, which was shown to be

localized to the periplasmic space and to be linked to AcrB and TolC (Kawabe et al.,

2000; Ma et al., 1993; Zgurskaya and Nikaido 1999). This complex forms an active

multidrug efflux machinery with a wide substrate range (for review see Nikaido 2000;

Nikaido and Zgurskaya 2001; Zgurskaya and Nikaido 2000). In contrast to the

situation in E. coli, all three components of this efflux pump of C. jejuni seem to be

encoded by the same gene cluster (Cj0365c-Cj0367c in C. jejuni NCTC 11168

genome sequence). These genes are transcribed in the same orientation on the

complementary strand, with acrA sharing the same promoter as the two other genes in

the operon, Cj0366 (multidrug efflux pump CmeB) and Cj0365 (putative outer

membrane channel protein). CmeB is a RND-type multi-substrate efflux transporter

that contributes to intrinsic resistance to a range of structurally unrelated compounds

in C. jejuni (Pumbwe and Piddock 2002).

AcrA of E. coli contains a cleavable signal peptide. After cleavage of the signal

peptide, the amino-terminal cysteine residue becomes acylated with palmitate

(Zgurskaya and Nikaido 1999). The same authors also showed that AcrA forms a

highly asymmetric protein capable to span the periplasm. A 2D crystal structure

supports this finding (Avila-Sakar et al., 2001). AcrA of C. jejuni also contains a

cleavable signal sequence and a putative prokaryotic membrane lipoprotein lipid

attachment site. AcrA could only be released from the membrane by addition of

detergent, supporting the lipid anchoring in Campylobacter as well (data not shown).

53

M K L F Q K N T I L A L G V V L L L A A C S K E E A P K I Q 1 AcrA(81-176)M K L F Q K N T I L A L G V V L L L T A C S K E E A P K I Q 1 AcrA(11168)

M P P Q P V T T M S A K S E D L P L S F T Y P A K L V S D Y 31 AcrA(81-176)M P P Q P V T T M S A K S E D L P L S F T Y P A K L V S D Y 31 AcrA(11168)

D V I I K P Q V S G V I V N K L F K A G D K V K K G Q T L F 61 AcrA(81-176)D V I I K P Q V S G V I E N K L F K A G D K V K K G Q T L F 61 AcrA(11168)

I I E Q D K F K A S V D S A Y G Q A L M A K A T F E N A S K 91 AcrA(81-176)I I E Q D K F K A S V D S A Y G Q A L M A K A T F E N A S K 91 AcrA(11168)

D F N R S K A L F S K S A I S Q K E Y D S S L A T F N N S K 121 AcrA(81-176)D F N R S K A L F S K S A I S Q K E Y D S S L A T F N N S K 121 AcrA(11168)

A S L A S A R A Q L A N A R I D L D H T E I K A P F D G T I 151 AcrA(81-176)A S L A S A R A Q L A N A R I D L D H T E I K A P F D G T I 151 AcrA(11168)

G D A L V N I G D Y V S A S T T E L V R V T N L N P I Y A D 181 AcrA(81-176)G D A L V N I G D Y V S A S T T E L V R V T N L N P I Y A D 181 AcrA(11168)

F F I S D T D K L N L V R N T Q S G K W D L D S I H A N L N 211 AcrA(81-176)F F I S D T D K L N L V R N T Q N G K W D L D S I H A N L N 211 AcrA(11168)

L N G E T V Q G K L Y F I D S V I D A N S G T V K A K A V F 241 AcrA(81-176)L N G E T V Q G K L Y F I D S V I D A N S G T V K A K A I F 241 AcrA(11168)

D N N N S T L L P G A F A T I T S E G F I Q K N G F K V P Q 271 AcrA(81-176)D N N N S T L L P G A F A T I T S E G F I Q K N G F K V P Q 271 AcrA(11168)

I G V K Q D Q N D V Y V L L V K N G K V E K S S V H I S Y Q 301 AcrA(81-176)I A V K Q N Q N D V Y V L L V K N G K V E K S S V H I S Y Q 301 AcrA(11168)

N N E Y A I I D K G L Q N G D K I I L D N F K K I Q V G S E 331 AcrA(81-176)N N E Y A I I D K G L Q N G D K I I L D N F K K I Q V G S E 331 AcrA(11168)

V K E I G A Q 361 AcrA(81-176)V K E I G A Q 361 AcrA(11168)

Figure 2.10: Alignment of deduced amino acid sequences of AcrA of two C. jejuni strains. Theputative glycoprotein AcrA of C. jejuni 81-176 was purified and identified. There are 6 amino acidexchanges between AcrA of C. jejuni 81-176 (top sequence) and C. jejuni NCTC 11168 (bottomsequence, GenBank accession number AL139075). Amino acids that are exchanged are marked.

OmpH1, like AcrA, contains consensus sequences for N-linked protein glycosylation.

The two sites with the sequence Asn-X-Ser/Thr are underlined in Table II. One

peptide containing a consensus sequence was found to be unmodified based on the

results from mass spectrometry. The peptide containing the second glycosylation site

was not found by mass spectrometry. Therefore, this consensus sequence was

postulated to be modified in vivo with a N-linked oligosaccharide.

OmpH1 of C. jejuni 81-176 was sequenced and the gene product compared to protein

sequence of OmpH1 of C. jejuni NCTC 11168 (Parkhill et al., 2000), A74/O

(Meinersmann et al., 1997) and 72Dz/92 strains (Pawelec et al., 2000) (Figure 2.11).

The first three strains showed almost the same sequence, whereas the latter showed a

bigger divergence. The exchanges are distributed uniformly along the protein.

Pawelec and coworkers showed that immunogenic proteins of C. jejuni like OmpH1

54

are genetically diverse (Pawelec et al., 2000). OmpH1 of C. jejuni 81-176 strain has a

calculated relative molecular weight of 30,969 Da, an isoelectric point (pI) of 5.69

and it contains a putative cleavable signal sequence. Database similarity searches with

the amino acid sequence of the mature protein showed significant levels of similarity

(34.5%) to the glutamine binding protein (GlnH) of Bacillus stearothermophilus (Wu

and Welker 1991). This protein is a periplasmic protein that is involved in glutamine

transport. However, involvement of OmpH1 of C. jejuni in glutamine transport was

not shown; the function of OmpH1 still remains unclear (Pawelec et al., 2000).

55

M K K I L L S V L T T F V A V V L A A C G G N S D S K T L N1 OmpH1(81-176)M K K I L L S V L T A F V A V V L A A C G G N S D S K T L N1 OmpH1(11168)M K K I L L S V L T T F V A V V L A A C G G N S D S K T L N1 OmpH1(A74)M K K M L L S I F T T F V A V F L A A C G G N S D S G A S N1 OmpH1(72Dz)

S L D K I K Q N G V V R I G V F G D K P P F G Y V D E K G N31 OmpH1(81-176)S L D K I K Q N G V V R I G V F G D K P P F G Y V D E K G N31 OmpH1(11168)S L D K I K Q N G V V R I G V F G D K P P F G Y V D E K G N31 OmpH1(A74)S L E R I K Q D G V V R I G V F G D K P P F G Y V D E K G V31 OmpH1(72Dz)

N Q G Y D I A L A K R I A K E L F G D E N K V Q F V L V E A61 OmpH1(81-176)N Q G Y D I A L A K R I A K E L F G D E N K V Q F V L V E A61 OmpH1(11168)N Q G Y D I A L A K R I A K E L F G D E N K V Q F V L V E A61 OmpH1(A74)N Q G Y D I V L A K R I A K E L L G D E N K V Q F V L V E A61 OmpH1(72Dz)

A N R V E F L K S N K V D I I L A N F T Q T P E R A E Q V D91 OmpH1(81-176)A N R V E F L K S N K V D I I L A N F T Q T P Q R A E Q V D91 OmpH1(11168)A N R V E F L K S N K V D I I L A N F T Q T P E R A E Q V D91 OmpH1(A74)A N R V E F L K S N K V D I I L A N F T Q T P E R A E Q V D91 OmpH1(72Dz)

F C L P Y M K V A L G V A V P K D S N I T S V E D L K D K T121 OmpH1(81-176)F C S P Y M K V A L G V A V P K D S N I T S V E D L K D K T121 OmpH1(11168)F C L P Y M K V A L G V A V P K D S N I T S V E D L I D K T121 OmpH1(A74)F C L P Y M K V A L G V A V P Q D S N I S S I E D L K D K T121 OmpH1(72Dz)

L L L N K G T T A D A Y F T Q D Y P N I K T L K Y D Q N T E151 OmpH1(81-176)L L L N K G T T A D A Y F T Q N Y P N I K T L K Y D Q N T E151 OmpH1(11168)L L L N K G T T A D A Y F T Q D Y P N I K T L K Y D Q N T E151 OmpH1(A74)L L L N K G T T A D A Y F T K E Y P D I K T L K Y D Q N T E151 OmpH1(72Dz)

T F A A L M D K R G D A L S H D N T L L F A W V K D H P D F181 OmpH1(81-176)T F A A L M D K R G D A L S H D N T L L F A W V K D H P D F181 OmpH1(11168)T F A A L M D K R G D A L S H D N T L L F A W V K D H P D F181 OmpH1(A74)T F A A L I D Q R G D A L S H D N T L L F A W V K E H P E F181 OmpH1(72Dz)

K M G I K E L G N K D V I A P A V K K G D K E L K E F I D N211 OmpH1(81-176)K M G I K E L G N K D V I A P A V K K G D K E L K E F I D N211 OmpH1(11168)K M G I K E L G N K D V I A P A V K K G D K E L K E F I D N211 OmpH1(A74)K M A I K E L G N K D V I A P A V K K G D K E L K E F I D N211 OmpH1(72Dz)

L I I K L G Q E Q F F H K A Y D E T L K A H F G D D V K A D241 OmpH1(81-176)L I I K L G Q E Q F F H K A Y D E T L K A H F G D D V K A D241 OmpH1(11168)L I I K L G Q E Q F F H K A Y D E T L K A H F G D D V K A D241 OmpH1(A74)L I T K L G E E Q F F H K A Y D E T L K S H F G D D V K A D241 OmpH1(72Dz)

D V V I E G G K I271 OmpH1(81-176)D V V I E G G K I271 OmpH1(11168)D V V I E G G K I271 OmpH1(A74)D V V I E G G K I271 OmpH1(72Dz)

M K K I L L S V L T T F V A V V L A A C G G N S D S K T L N1 OmpH1(81-176)M K K I L L S V L T A F V A V V L A A C G G N S D S K T L N1 OmpH1(11168)M K K I L L S V L T T F V A V V L A A C G G N S D S K T L N1 OmpH1(A74)M K K M L L S I F T T F V A V F L A A C G G N S D S G A S N1 OmpH1(72Dz)

S L D K I K Q N G V V R I G V F G D K P P F G Y V D E K G N31 OmpH1(81-176)S L D K I K Q N G V V R I G V F G D K P P F G Y V D E

M K K I L L S V L T T F V A V V L A A C G G N S D S K T L N1 OmpH1(81-176)M K K I L L S V L T A F V A V V L A A C G G N S D S K T L N1 OmpH1(11168)M K K I L L S V L T T F V A V V L A A C G G N S D S K T L N1 OmpH1(A74)M K K M L L S I F T T F V A V F L A A C G G N S D S G A S N1 OmpH1(72Dz)

S L D K I K Q N G V V R I G V F G D K P P F G Y V D E K G N31 OmpH1(81-176)S L D K I K Q N G V V R I G V F G D K P P F G Y V D E K G N31 OmpH1(11168)S L D K I K Q N G V V R I G V F G D K P P F G Y V D E K G N31 OmpH1(A74)S L E R I K Q D G V V R I G V F G D K P P F G Y V D E K G V31 OmpH1(72Dz)

N Q G Y D I A L A K R I A K E L F G D E N K V Q F V L V E A61 OmpH1(81-176)N Q G Y D I A L A K R I A K E L F G D E N K V Q F V L V E A61 OmpH1(11168)N Q G Y D I A L A K R I A K E L F G D E N K V Q F V L V E A61 OmpH1(A74)N Q G Y D I V L A K R I A K E L L G D E N K V Q F V L

K G N31 OmpH1(11168)S L D K I K Q N G V V R I G V F G D K P P F G Y V D E K G N31 OmpH1(A74)S L E R I K Q D G V V R I G V F G D K P P F G Y V D E K G V31 OmpH1(72Dz)

N Q G Y D I A L A K R I A K E L F G D E N K V Q F V L V E A61 OmpH1(81-176)N Q G Y D I A L A K R I A K E L F G D E N K V Q F V L V E A61 OmpH1(11168)N Q G Y D I A L A K R I A K E L F G D E N K V Q F V L V E A61 OmpH1(A74)N Q G Y D I V L A K R I A K E L L G D E N K V Q F V L V E A61 OmpH1(72Dz)

A N R V E F L K S N K V D I I L A N F T Q T P E R A E Q V D91 OmpH1(81-176)A N R V E F L K S N K V D I I L A N F T Q T P Q R A E Q V D91 OmpH1(11168)A N R V E F L K S N K V D I I L A N F T Q T P E R A E Q V D91 OmpH1(A74)A N R V E F L K S N K V D I I L A N F T Q T P E R A E Q V D91 OmpH1(72Dz)

F C L P Y M K V A L G V A V P K D S N I T S V E D L K D K T121 OmpH1(81-176)F C S P Y M K V A L G V A V P K D S N I T S V

V E A61 OmpH1(72Dz)

A N R V E F L K S N K V D I I L A N F T Q T P E R A E Q V D91 OmpH1(81-176)A N R V E F L K S N K V D I I L A N F T Q T P Q R A E Q V D91 OmpH1(11168)A N R V E F L K S N K V D I I L A N F T Q T P E R A E Q V D91 OmpH1(A74)A N R V E F L K S N K V D I I L A N F T Q T P E R A E Q V D91 OmpH1(72Dz)

F C L P Y M K V A L G V A V P K D S N I T S V E D L K D K T121 OmpH1(81-176)F C S P Y M K V A L G V A V P K D S N I T S V E D L K D K T121 OmpH1(11168)F C L P Y M K V A L G V A V P K D S N I T S V E D L I D K T121 OmpH1(A74)F C L P Y M K V A L G V A V P Q D S N I S S I E D L K D K T121 OmpH1(72Dz)

L L L N K G T T A D A Y F T Q D Y P N I K T L K Y D Q N T E151 OmpH1(81-176)L L L N K G T T A D A Y F T Q N Y P N I K T L K Y D Q N T E151 OmpH1(11168)L L L N K G T T A D A Y F T Q D Y P N I K T L K Y D Q N T E151 OmpH1(A74)L L L N K G T T A D A Y F T K E Y P D I K T

E D L K D K T121 OmpH1(11168)F C L P Y M K V A L G V A V P K D S N I T S V E D L I D K T121 OmpH1(A74)F C L P Y M K V A L G V A V P Q D S N I S S I E D L K D K T121 OmpH1(72Dz)

L L L N K G T T A D A Y F T Q D Y P N I K T L K Y D Q N T E151 OmpH1(81-176)L L L N K G T T A D A Y F T Q N Y P N I K T L K Y D Q N T E151 OmpH1(11168)L L L N K G T T A D A Y F T Q D Y P N I K T L K Y D Q N T E151 OmpH1(A74)L L L N K G T T A D A Y F T K E Y P D I K T L K Y D Q N T E151 OmpH1(72Dz)

T F A A L M D K R G D A L S H D N T L L F A W V K D H P D F181 OmpH1(81-176)T F A A L M D K R G D A L S H D N T L L F A W V K D H P D F181 OmpH1(11168)T F A A L M D K R G D A L S H D N T L L F A W V K D H P D F181 OmpH1(A74)T F A A L I D Q R G D A L S H D N T L L F A W V K E H P E F181 OmpH1(72Dz)

K M G I K E L G N K D V I A P A V K K G D K E L K E F I D N211 OmpH1(81-176)K M G I K E L G N K D V I A P A V K K

L K Y D Q N T E151 OmpH1(72Dz)

T F A A L M D K R G D A L S H D N T L L F A W V K D H P D F181 OmpH1(81-176)T F A A L M D K R G D A L S H D N T L L F A W V K D H P D F181 OmpH1(11168)T F A A L M D K R G D A L S H D N T L L F A W V K D H P D F181 OmpH1(A74)T F A A L I D Q R G D A L S H D N T L L F A W V K E H P E F181 OmpH1(72Dz)

K M G I K E L G N K D V I A P A V K K G D K E L K E F I D N211 OmpH1(81-176)K M G I K E L G N K D V I A P A V K K G D K E L K E F I D N211 OmpH1(11168)K M G I K E L G N K D V I A P A V K K G D K E L K E F I D N211 OmpH1(A74)K M A I K E L G N K D V I A P A V K K G D K E L K E F I D N211 OmpH1(72Dz)

L I I K L G Q E Q F F H K A Y D E T L K A H F G D D V K A D241 OmpH1(81-176)L I I K L G Q E Q F F H K A Y D E T L K A H F G D D V K A D241 OmpH1(11168)L I I K L G Q E Q F F H K A Y D E T L K A H F G D D V K A D241 OmpH1(A74)L I T K L G E E Q F F H K A Y D E T

G D K E L K E F I D N211 OmpH1(11168)K M G I K E L G N K D V I A P A V K K G D K E L K E F I D N211 OmpH1(A74)K M A I K E L G N K D V I A P A V K K G D K E L K E F I D N211 OmpH1(72Dz)

L I I K L G Q E Q F F H K A Y D E T L K A H F G D D V K A D241 OmpH1(81-176)L I I K L G Q E Q F F H K A Y D E T L K A H F G D D V K A D241 OmpH1(11168)L I I K L G Q E Q F F H K A Y D E T L K A H F G D D V K A D241 OmpH1(A74)L I T K L G E E Q F F H K A Y D E T L K S H F G D D V K A D241 OmpH1(72Dz)

D V V I E G G K I271 OmpH1(81-176)D V V I E G G K I271 OmpH1(11168)D V V I E G G K I271 OmpH1(A74)D V V I E G G K I271 OmpH1(72Dz)

Figure 2.11: Alignment of deduced amino acid sequences of OmpH1 of C. jejuni strains. Theputative glycoprotein OmpH1 of C. jejuni 81-176 was purified and identified. The genomic DNA wassequenced and the deduced amino acid compared to the OmpH1 sequences of C. jejuni strains NCTC11168 (accession number CAB73238), A74/O (accession number U93169) and 72Dz/92 (GenBankaccession number Y10872). Amino acids that are exchanged are marked.

56

2.2.6 Deletion of acrA and ompH1 and complementation in trans

To prove that AcrA and OmpH1 are indeed the proteins carrying the pgl-dependent

modification detected by the antiserum, the two genes encoding these proteins were

deleted in the C. jejuni genome by insertion of a Kmr cassette. Membrane proteins of

these mutant cells were separated by SDS-PAGE and stained by silver (Figure 2.12A)

or transferred to nitrocellulose membrane and incubated with the polyclonal antiserum

(Figure 2.12B). As shown before, the 47 and 35 kDa proteins were absent in extracts

derived from pglB mutant cells (Figure 2.12B, lane 2, indicated by arrows). This

deficiency was complemented by expression of the wild-type allele (lane 3). In acrA

and ompH1 mutant cells, the 47 or 35 kDa proteins were absent as visualized by

Western blots using the antiserum (lane 4 and 6, respectively). However, the

expression of these proteins was restored by introduction of plasmids encoding AcrA

or OmpH1 (lane 5 and 7).

57

B

7654321

47.5

32.5

175

83

62

B

7654321

47.5

32.5

175

83

62

7654321 7654321

47.5

32.5

175

83

62

47.5

32.5

175

83

62

47.5

32.5

175

83

62

AkDa

47.5

32.5

175

83

62

kDa

47.5

32.5

175

83

62

kDa

47.5

32.5

175

83

62

47.5

32.5

175

83

62

WT

pglB-pglB-

p(pg

l)

pom

pH1

acrA-acrA- ompH1-ompH1-

pacr

A

Figure 2.12: Analysis of membrane proteins of C. jejuni WT, pglB, acrA and ompH1 mutants andcomplementation in trans. Membrane proteins were separated by SDS-PAGE and either stained bysilver (A) or after blotting immunodetected with the C. jejuni antiserum (B). Lane 1, WT; lane 2, pglBmutant; lane 3, pglB mutant containing a plasmid that expresses PglB; lane 4, acrA mutant; lane 5,acrA mutant containing a plasmid expressing AcrA; lane 6, ompH1 mutant; lane 7, ompH1 mutantcontaining a plasmid expressing OmpH1. Molecular weight markers are indicated on the left. Thearrows indicate the position of the two glycoproteins that were absent in the mutant strains but werecomplemented in trans by introduction of the specific plasmids.

58

2.2.7 Expression of AcrA in E. coli in the presence of the pgl

locus leads to glycosylation

The identification of target proteins for the putative N-linked protein glycosylation

process made it possible to verify the role of the pgl locus in this process by

reconstituting protein glycosylation in the non-glycosylating model-organism E. coli.

Therefore, AcrA was expressed in E. coli under the control of an inducible promoter

in the presence or absence of the pgl locus. In addition the pgl locus harboring a pglB-

mutant allele with the unreative W458A, D459A mutations served as a control.

AcrA was amplified by PCR, cloned into pET24 to achieve controlled expression of

the protein and transformed into E. coli BL21 cells. Membrane proteins of these cells

were analyzed by SDS-PAGE and visualized by Coomassie Blue staining (Figure

2.13A). When T7 RNA polymerase was induced in the host cells by addition of IPTG,

AcrA was expressed in high levels as shown by the presence of a protein migrating at

around 45 kDa (Figure 2.13A, lane 4). Overexpression of this protein was also

observed in cells that simultaneously harbored the pgl locus (lane 5 and 6,

respectively).

To visualize potential glycosylation of the AcrA protein, the glycosylation specific

antiserum was used in Western blot experiments (Figure 2.13B). When AcrA was

expressed in the presence of the wild-type version of the pgl locus, the antiserum

recognized two proteins, migrating at 46 kDa and 47 kDa (lane 5). These proteins

were absent when AcrA was expressed in the presence of PglB with two amino acid

substitutions (lane 6) and or in the absence of the pgl locus (lane 4). In addition, these

two proteins were not detected in cells not expressing AcrA but only the pgl locus

(lane 1-3). It was concluded that the pgl locus directs glycosylation of AcrA in E. coli

and that PglB is essential for this process.

To address this observation directly, the SBA lectin, known to bind specifically

terminal GalNAc residues and to bind C. jejuni glycoproteins ((Linton et al., 2002),

Figure 2.9B), was used for the analysis. In case of a complete, C. jejuni-type

glycosylation of AcrA in E. coli, cross-reacting proteins to SBA lectin have to be

expected in an acrA- and pgl-dependent manner. Indeed, the results obtained are

identical to those observed with the antiserum specific for glycoproteins (Figure

2.13C). When AcrA was expressed in the presence of the pgl locus, two proteins were

recognized by the lectin (lane 5). These proteins were not seen when PglB with two

59

amino acids substitutions was expressed (lane 6) or when AcrA was expressed in the

absence of the locus (lane 4). These glycoproteins had the same mobility in SDS-

PAGE as the AcrA protein detected by the glycosylation-specific antiserum. In

contrast to the serum, SBA lectin seems to have a weak affinity for the highly

expressed protein, resulting in the faint staining of this protein.

The presence of two bands representing glycosylated AcrA protein in E. coli suggests

that different glycoforms are produced, containing either one or two protein-bound

oligosaccharide(s). To prove that AcrA glycoproteins were produced in E. coli,

modified AcrA was purified and the oligosaccharide composition of this protein was

determined. In addition, these experiments allowed a direct assessment of the

glycosylation site of the AcrA protein. One glycosylation site was identified in

collaboration with A. Dell and coworkers (Imperial College, London). Using mass

spectrometry, they showed that the peptide DFNR is modified with an

oligosaccharide. This confirmed that AcrA contains N-linked oligosaccharides and

that one of the proposed consensus sequences (Asn-Arg-Ser, Table II) was

glycosylated. Therefore, the acceptor for N-linked protein glycosylation in C. jejuni is

postulated to be the same as in eukaryotes and archaea. However the composition of

the oligosaccharide differs from the N-linked oligosaccharide of eukaryotes. A

preliminary structure is depicted in Figure 2.14. An unusual 2,4-diacetimido-2,4,6-

trideoxyhexose (DATDH) is linked to the amide nitrogen of Asn. Nothing is known

about the linkages of the monosaccharides, except that the terminal GalNAc is linked

via an α1-3 bond to either GalNAc or GlcNAc. More experiments have to be done to

resolve the final structure of the oligosaccharide.

60

A

B

47.5

32.5

175

83

62

B

47.5

32.5

175

83

62

47.5

32.5

175

83

62

47.5

32.5

175

83

62

47.5

32.5

175

83

62

7654321

C

47.5

32.5

175

83

62

7654321

47.5

32.5

175

83

62

47.5

32.5

175

83

62

47.5

32.5

175

83

62

7654321 7654321

C

47.5

32.5

175

83

62

kDa

47.5

32.5

175

83

62

kDa

47.5

32.5

175

83

62

47.5

32.5

175

83

62

kDa

pET24b pET24(AcrA)

E. coli BL21 membranes

pAC

YC

184

pAC

YC

(pgl

)

p(W

458 A

, D45

9A)

p AC

YC

184

pAC

YC

(pgl

)

p(W

458 A

, D45

9A)

Figure 2.13: Analysis of membrane proteins of E. coli BL21(DE3) cells overexpressing AcrA inthe presence or absence of the pgl locus. Membrane proteins were separated by SDS-PAGE andeither stained by Coomassie Blue (A) or, after transfer to nitrocellulose, immunodetected with the C.jejuni antiserum (B) or probed with SBA lectin (C). Lane 1-3, cells containing the plasmid backbone(pET24b) only; lane 4-6, cells overexpressing AcrA of C. jejuni. As a second plasmid these cellscontained pACYC184 (lane 1 + 4), pACYC184 with the pgl locus (lane 2 + 5) or pACYC184 with thepgl locus expressing PglB with amino acid substitutions (W458A and D459A) in the highly conservedmotif (lane 3 + 6). For comparison a purified glycoprotein fraction of solubilized membrane proteins ofC. jejuni was applied in lane 7. Molecular weight markers are indicated on the left.

61

GalNAc-HexNAc-HexNAc-HexNAc-HexNAc-DATDH

O

O

O

OH

OH

NAc

HO

HO

O

O

O

O

O

HO

O

O

O

HO

AcN

NAc

NAc

HO

NAc

NAc

HO

NAC

HO

OH

O

O

HO

HO

OH

DFNR

O

O

O

OH

OH

NAc

HO

HO

O

O

O

O

O

HO

O

O

O

HO

AcN

NAc

NAc

HO

NAc

NAc

HO

NAC

HO

OH

O

O

HO

HO

OH

DFNR

Hex

Figure 2.14: Preliminary structure of the N-linked oligosaccharide in C. jejuni. Theoligosaccharide is linked to the amide nitrogen of Asn of AcrA within Asn-Arg-Ser. The structure andlinkages of the monosaccharides are not known, except an unusual 2,4-diacetimido-2,4,6-trideoxyhexose (DATDH) at the reducing end and a terminal GalNAc(α1-3).

62

2.3 Discussion

2.3.1 Pgl locus encodes N-linked protein glycosylation pathway

of C. jejuni

It was shown that the pgl locus is responsible for production of N-linked

glycoproteins in C. jejuni. Two glycoproteins, AcrA and OmpH1, were identified.

The structure analysis (Figure 2.14) revealed that the composition of the

oligosaccharide does not show similarities to the highly conserved structure of the

oligosaccharide assembled on dolichyl pyrophosphate at the ER membrane of

eukaryotes. However, the oligosaccharide is linked to the amide nitrogen of Asn

within the consensus sequence Asn-X-Ser/Thr of AcrA. This sequence is necessary

for N-linked protein glycosylation in eukaryotes (Figure 1.2). The presence of this

consensus sequence on glycoproteins of C. jejuni suggests that the Ser or Thr residues

are essential for glycosylation in C. jejuni as well. However, mutagenesis of these

residues has to be done to verify this hypothesis.

AcrA contains two N-linked oligosaccharides, whereas OmpH1 contains one N-linked

oligosaccharide, most likely linked to Asn of Asn-Ile-Thr; however this has not been

shown experimentally. There was no cross-reactivity of the SBA lectin to OmpH1 in

pglB mutant (data not shown). SBA binds specifically terminal GalNAc, therefore it is

proposed that the same oligosaccharide linked to AcrA is linked to OmpH1.

Interestingly, OmpH1 was identified as an antigen in two different C. jejuni strains

(Meinersmann et al., 1997). Western blot analysis revealed antigenic differences in

this protein of the two strains. In one strain, able to colonize chicken, antigen

expressed had a size of 34 kDa. In the other strain, being a poor colonizer, the

antigens expressed had a size of 32 and 34 kDa. The authors had shown that these two

strains do not show a difference of OmpH1 sequence. The finding that OmpH1 is

glycosylated in a PglB dependent way may explain these antigenic differences of the

two strains. Moreover, it shows that glycosylation in C. jejuni could play an important

role in colonization. Indeed, C. jejuni pglB mutant showed a reduced ability of

adhesion and invasion in vitro and an almost complete loss of colonization of the

intestinal tracts of mice in vivo (Szymanski et al., 2002).

Glycoproteins of C. jejuni are low in abundance, but are highly immunoreactive

proteins with antibodies raised against glycan rather than amino acid epitopes (Figure

63

2.5). In addition to AcrA and OmpH1, two other glycoproteins had been isolated from

C. jejuni (Linton et al., 2002). PEB3 and CgpA were also shown to be

immunoreactive proteins. The deletion of pglB led to a loss of glycosylation of PEB3

(data not shown). Moreover, the same oligosaccharide as in AcrA is attached to PEB3

(B. Wren, personal communication). Both proteins contain various N-glycosylation

sites and therefore are expected to contain N-linked oligosaccharides. It is evident that

C. jejuni is able to produce different N-linked glycoproteins, as additional

immunoreactive proteins were eluted from the SBA column (Figure 2.9, lane 7).

Interestingly, another glycoprotein, flagellin, contains up to 19 pseudoaminic acid

residues, which are covalently linked either to Ser or Thr residues (Thibault et al.,

2001). However, the pgl locus is not involved in the glycosylation of this protein, but

instead, genes located in the same cluster as the genes encoding the flagellin structural

subunits FlaA and FlaB are responsible for glycosylation of flagellin (Guerry et al.,

1996; Linton et al., 2000; Thibault et al., 2001).

2.3.1.1 PglB plays a major role in the glycosylation of proteins in C.

jejuni

The membrane topology of PglB and its eukaryotic and archaeal homologs is highly

conserved. The N-terminal domain is hydrophobic whereas the C-terminal domain is

hydrophilic. The membrane topology of PglB was studied in the plasma membrane of

E. coli. It was concluded that the hydrophilic C-terminus is located in the periplasmic

space (Figure 2.6), the equivalent to the ER of eukaryotes. The hydrophilic C-

terminus of Stt3p, the S. cerevisiae homolog of PglB, is located in the lumen of the

ER (Zufferey et al., 1995), where the transfer of the oligosaccharide from the carrier

lipid to the protein takes place during N-linked protein glycosylation. A similar

transfer of an oligosaccharide from the carrier lipid polyisoprenyl pyrophosphate

takes place in the periplasmic space during LPS biosynthesis of bacteria. However,

the acceptor is not a protein but instead the lipid A anchor (Raetz 1996; Whitfield

1995).

PglB is not involved in the biosynthesis of LPS (data not shown). Therefore it was

postulated to be involved in protein glycosylation. A change in the recognition of

proteins by the antiserum was observed in the pglB mutant. The WT pattern could be

restored by the expression of the wild-type pglB allele, but only partially by

expression of mutant alleles containing point mutations in the highly conserved motif

64

(Figure 2.7). Therefore, this highly conserved motif, located in the periplasmic space,

is crucial for glycosylation. Amino acid exchanges in this motif led to

underglycosylation or complete loss of glycosylation in C. jejuni. This provides

further evidence that PglB functions as an OTase and transfers the oligosaccharide to

proteins. AcrA lacking a signal sequence expressed in E. coli in the presence of the

pgl locus was not glycoslyated (data not shown). Therefore, N-linked protein

glycosylation takes place in the periplasmic space.

Protein underglycosylation was also observed in pglE (Figure 2.8) and pglF (data not

shown) mutants; however the phenotype is not as severe as in the pglB mutant. A

similar observation has been made in the eukaryotic system. Mutations affecting the

biosynthesis of the LLO are less severe than mutations of the OTase (Burda et al.,

1996; Burda and Aebi 1998; Burda et al., 1999; Jakob et al., 1998). This is due to the

fact that the OTase exhibits reduced affinity for incompletely assembled substrates.

The similarity of the phenotype observed in eukaryotes and C. jejuni supports the

model that the N-protein glycosylation pathway in C. jejuni shares significant

similarities to the eukaryotic pathway.

65

2.3.1.2 Potential pathway for N-linked protein glycosylation in C. jejuni

Based on the oligosaccharide structure (Figure 2.14), the following model for the

pathway of N-linked protein glycosylation in C. jejuni (see Figure 2.4) is postulated

(Figure 2.15).

PglB

PglC PglAHIJ

Cytoplasm

Periplasm

PP

PP

Asn

PPPWlaB

dTDP- dTMP

galE wlaB pglH pglI pglJ pglB pglA pglC pglD pglE pglF pglGgalE wlaB pglH pglI pglJ pglB pglA pglC pglD pglE pglF pglG

= Bactoprenol= Bactoprenol

= DATDH

= HexNAc

= GalNAc

= Hex

= DATDH= DATDH

= HexNAc= HexNAc

= GalNAc= GalNAc

= Hex= Hex

Figure 2.15: Proposed model for N-linked protein glycosylation pathway in C. jejuni. Theoligosaccharide (Figure 2.14) is synthesized in a stepwise assembly on the carrier lipid undecaprenylpyrophosphate at the cytoplasmic side of the plasma membrane. PglF, PglE and PglD most likely areinvolved in the synthesis of the nucleotide sugar dTDP-DATDH. PglC initiates the assembly of theLLO by transferring this DATDH-P to undecaprenyl phosphate. The glycosyltransferase homologsPglA, PglI, PglH and PglJ complete the synthesis of the LLO at the cytoplasmic side of the plasmamembrane. WlaB, an ABC transporter, may translocate the LLO into the periplasmic space. In thecentral reaction of the process, PglB, the OTase, transfers the oligosaccharide to proteins.

No direct evidence has been obtained that the oligosaccharide is assembled on

undecaprenyl pyrophosphate. However, the introduction of the pgl locus into E. coli

cells is sufficient for glycosylation of AcrA (Figure 2.13B/C). Since there are no

homologs to proteins involved in lipid biosynthesis encoded by the pgl locus, the

carrier lipid has to be synthesized by E. coli. Most likely undecaprenyl pyrophosphate

is used for the assembly of the oligosaccharide as it is involved in the synthesis of the

oligosaccharide unit of extracellular polysaccharides. When AcrA was overexpressed

66

in E. coli, it was not completely glycosylated like in C. jejuni, but instead most of the

protein was unglycosylated (Figure 2.13). Most likely the synthesis of the LLO is the

rate limiting factor.

PglC shows homologies to proteins that are involved in the synthesis of bacterial cell

walls and surface polysaccharides. They catalyze the formation of the undecaprenyl

pyrophosphate monosaccharide bond by transferring the monosaccharide-phosphate

from an activated nucleotide precursor to the carrier lipid undecaprenyl phosphate

(Bugg and Brandish 1994). For PglC, the highest similarity was found to RfbP of

Salmonella typhimurium (Fry et al., 1998), which transfers Gal-P to undecaprenyl

phosphate in the first step of O-polysaccharide biosynthesis (Whitfield 1995). In C.

jejuni, the oligosaccharide is linked via the unusual 2,4-diacetimido-2,4,6-

trideoxyhexose (DATDH) to the amide nitrogen of Asn (Figure 2.14). Therefore it is

postulated that PglC initiates the LLO biosynthesis by transferring DATDH to the

lipid using dTDP-DATDH as a substrate. DATDH is covalently linked to pili of

Neisseria meningitidis (Stimson et al., 1995). The pgl (protein glycosylation) operon

of N. meningitidis is responsible for glycosylation of the pili (Power et al., 2000).

PglD, pglE and pglF of the pgl locus of C. jejuni show high homologies to pglB, pglC

and pglD of the pgl locus of N. meningitidis. The proteins encoded by these genes are

postulated to be involved in the synthesis of dTDP-DATDH. The biosynthesis of 4-

acetamido-4,6-dideoxyhexose had been described in E. coli and is initiated by

dehydration (PglF dependent), followed by transmutation (PglE dependent) and

transacetylation (PglD dependent) of TDP-Glc (Dietzler and Strominger 1973;

Matsuhashi and Strominger 1966).

The striking difference of the pgl operon of N. meningitidis as compared to the pgl

operon of C. jejuni is the absence of a pglB homolog in N. meningitidis. Therefore,

DATDH is probably transferred directly from dTDP to the hydroxylgroup of Ser 63,

whereas PglC of C. jejuni transfers DATDH-P from dTDP to undecaprenyl

phosphate. PglE (Figure 2.8) and pglF (data not shown) C. jejuni mutants showed

reduced glycosylation of AcrA, suggesting that PglC has reduced affinity for the

precursor of dTDP-DATDH and transfers this monosaccharide-P to the carrier lipid.

Alternatively, the formation of DATDH can take place on the lipid-bound

oligosaccharide.

Most likely the glycosyltransferase homologs PglA, PglI, PglH and PglJ are

responsible for completion of full length LLO at the cytoplasmic side of the plasma

67

membrane. GalE was identified as a UDP-glucose-4-epimerase, which catalyzes the

interconversion of UDP-glucose and UDP-galactose (Fry et al., 2000). These authors

showed that the galE mutant synthesize a truncated LPS molecule indicating a role of

GalE in biosynthesis of the core of LPS. The effect on protein glycosylation was not

examined, but GalE could also be involved in protein glycosylation, providing a

nucleotide sugar precursor for the synthesis of the LLO.

WlaB shows homologies to ABC transporters. MsbA, the E. coli homolog, is involved

in transferring the lipid A core molecule from the cytoplasmic side of the plasma

membrane to the periplasmic side (Zhou et al., 1998). MsbA is the only essential

ABC transporter in E. coli (Zhou et al., 1998) and was crystallized recently (Chang

and Roth 2001). WlaB of C. jejuni may translocate the LLO to the periplasmic space.

It had been suspected that deletion of wlaB leads to cell death of C. jejuni (Fry et al.,

2000). This is most likely due to another function of WlaB than its role in protein

glycosylation. Protein glycosylation seems to be not essential for C. jejuni, as pglB

deleted cells are viable. WlaB might be involved in the translocation of the lipid A

anchor to the periplasmic space as well.

After translocation of the LLO into the periplasmic space, the putative OTase PglB

transfers the oligosaccharide from the carrier lipid to proteins with the consensus

sequence Asn-X-Ser/Thr. As there are no homologs of pglB found in the genome of

other bacteria, C. jejuni probably acquired pglB through horizontal gene transfer. This

enables the organism to produce N-glycoproteins that play an important role in C.

jejuni pathogenesis.

68

2.3.2 Production of N-glycoproteins in E. coli

N-linked glycoproteins were produced for the first time in E. coli. Mono- and

diglycosylated AcrA was expressed in the presence of the pgl locus (Figure 2.13). The

purification and analysis of the oligosaccharide revealed the same structure of the

oligosaccharide produced in E. coli as compared to endogenous oligosaccharide of

PEB3. Therefore, the N-linked protein glycosylation pathway (Figure 2.15) can be

reproduced and studied in E. coli. Single genes of the pgl locus will be deleted and the

structure of lipid-linked oligosaccharide and protein-linked oligosaccharide analyzed.

This will help to reveal the function of the enzymes encoded by genes of the pgl locus

of C. jejuni.

Further, the reconstitution of N-linked protein glycosylation of C. jejuni in E. coli may

help to express eukaryotic glycoproteins in E. coli. The inability of E. coli cells to

exert glycosylation of proteins has been the strongest drawback for its use as the

preferred host for the production of human proteins. Many human therapeutics are

glycoproteins, and the importance of the posttranslational modification of

polypeptides with defined oligosaccharides is well documented by their implication in

numerous biological phenomena. Since the majority of therapeutically relevant

proteins are glycosylated in their natural forms, they should also be glycosylated as

recombinant proteins in order to get the correct biological activity. Systems for the

expression of glycosylated proteins have been developed. The most commonly used

are Chinese hamster ovary (CHO) cell lines (Grabenhorst et al., 1999), insect cells

(Altmann et al., 1999) or fungal cells (Malissard et al., 1999). These cell lines have all

the capability to glycosylate proteins, an essential process in eukaryotic cells, but they

exhibit major differences in the production of recombinant glycoproteins. The

synthesis of the LLO and the transfer of the oligosaccharide to polypeptides is a

highly conserved mechanism in all eukaryotes (Figure 1.1), whereas further

processing and trimming of the N-glycans in the Golgi vary between organisms and

cell type. Therefore, the final structure of the recombinant glycoprotein is defined by

the production cell used, and all of the eukaryotic production systems exhibit this

specificity. It might be possible to genetically engineer eukaryotic production cell

lines in such a way that a defined oligosaccharide structure is produced. However, the

plethora of glycosyltransferases active in the Golgi compartment of eukaryotes makes

such an approach very difficult.

69

The protein glycosylation operon of C. jejuni allows E. coli to form an N-glycosidic

linkage at the Asn of the Asn-X-Ser/Thr consensus sequence. Since E. coli is easier to

handle and to grow and its genetics are very well known, the production of N-

glycoproteins in E. coli might represent a breakthrough in biotechnology. The

introduction of a protein glycosylation operon into an organism that normally does not

glycosylate proteins allows manipulating the structure of the N-glycan by modifying

the different glycosyltransferases. In contrast, specific oligosaccharide structure in

eukaryotic cells requires the tailoring of a highly complex, essential pathway, and this

might possibly interfere with the viability of the production cell.

In contrast to eukaryotes, the tailoring in E. coli can be obtained by the introduction of

specific components of the glycosylation machinery that lead to the desired

glycoprotein. The specific nucleotide sugar precursors that are required for the

assembly of eukaryotic oligosaccharides are present in E. coli. These precursors are

used for the biosynthesis of extracellular polysaccharides in E. coli. Therefore specific

glycosyltransferases have to be introduced into E. coli that are able to assemble the

required oligosaccharide on polyisoprenyl pyrophosphate. In addition the specificity

of the putative transporter WlaB and of the OTase PglB has to be relaxed. These two

enzymes have to be modified to recognize the substrate that differs from its native

composition. PglB is able to transfer incompletely assembled oligosaccharide in C.

jejuni (Figure 2.9); however it is not known whether it is able to transfer a completely

different oligosaccharide structure. The E. coli homolog of WlaB, MsbA is able to

transport incompletely assembled lipid A core through the inner membrane (Zhou et

al., 1998); however, nothing is known about the substrate specificity of the C. jejuni

homolog.

70

2.4 Material and methods

2.4.1 Bacterial strains and plasmids

Campylobacter jejuni 81-176 strain was originally isolated from a gastroenteritis

outbreak (Korlath et al., 1985) and its virulence was confirmed in human volunteers

(Black et al., 1988). Escherichia coli DH5α (supE44 ∆lac����� � ��� lacZ∆M15)

hsdR17 recA1 endA1 gyrA96 thi-1 relA1) was used as the host for cloning

experiments, E. coli CC118 (araD139 ∆(ara-leu)7697 ∆lacX74 ∆(phoA)20 galE galK

rpsL thi) as the host for PhoA studies and E. coli BL21 (DE3) (hsdS gal (λcIts857

ind1 Sam7 nin5 lacUV5-T7 gene 1)) as the host for high level expression of AcrA. E.

coli DH5α(RK212.1) was used as a donor in conjugation experiments (Figurski and

Helinski 1979). Plasmids pACYC184 (NEB, Beverly, MA, USA), pSP73 (Promega,

Wallisellen, CH), pET16b (Promega, Wallisellen, CH), pET24b (Promega,

Wallisellen, CH) and pRY111 (Yao et al., 1993) were used as cloning vectors.

2.4.2 Media and growth conditions

C. jejuni strain was routinely grown on solid Mueller-Hinton (MH) agar (Difco) under

microaerobic conditions (85% N2, 10% CO2 and 5% O2). E. coli strains were grown

on Luria-Bertani agar. When appropriate, antibiotics were added at the following

concentrations: kanamycin (Km), 50 µg/ml; chloramphenicol, 10 µg/ml; and

ampicillin, 100 µg/ml.

2.4.3 Protein electrophoresis and blotting

SDS-PAGE and Western blot analysis was performed as described (Aebi et al., 1996).

Antiserum against C. jejuni 81116 whole cell extracts was diluted 20,000 fold (D. G.

Newell). 0.25 µg/ml polyclonal rabbit IgG fractions α-PhoA (stock solution 5 mg/ml)

(NEB, Gilbertsville, PA, USA) was used. Detection of these antisera was performed

using goat-α-rabbit horseradish peroxidase (final concentration of 0.5 µg/ml, stock

solution 1 mg/ml) (Promega, Wallisellen, CH). Detection was done with the enhanced

Chemiluminescence ECL system (Amersham, Little Chalfont, GB). SBA horseradish

peroxidase was used for lectin blots (final concentration 1.3 µg/ml, stock solution 0.5

mg/ml) (Sigma-Aldrich, Buchs, CH).

71

2.4.4 Chemical deglycosylation

Membrane proteins were dialyzed against water, dried under vacuum and

deglycosylated with TFMS using the GlycoFree Deglycosylation Kit (Glyko Inc,

Novato, CA, USA) according to the instructions of the manufacturer.

2.4.5 Mutagenesis of C. jejuni genes

Mutagenesis of pglB, acrA and ompH1 was carried out by replacing most of the

coding sequence with a kanamycin resistance cassette. For deletion of pglB, the

following primers were used. 5’-GGGAGTACTTTTTAGGTGAAGGTGTGC-3’ and

5’-TATCGATCGCCACCAAAGATAAATTCC-3’ (for amplification of 5’

fragment), 5’-GCATATGCATATGATACGCCAAAAACTCGTG-3’ and 5’-

CCGGGTCTCGACCG-CCCTACGGATTTTATCAC-3’ (for amplification of 3’

fragment). For deletion of acrA, the following primer pairs were used: 5’-

GGGAGTACTAATAGTTGTTATCAGGGC-3’ and 5’-

TATCGATCGTTGTTACAGGTTGAGGCG-3’ (for amplification of 5’ fragment),

5’-GGATTAATCCAAAACAATGAATATGC-3’ and 5’-GGATTAATCCC-

CTGATTGAAGATTGATC-3’ (for amplification of 3’ fragment). For deletion of

ompH1, the following primer pairs were used: 5’-GGGAGTACTAGCTTGTT-

TATCCATAAC-3’ and 5’-TATCGATCGAGTCAGAATTTCCTCCAC-3’ (for

amplification of 5’ fragment), 5’-GGATTAATGCAGTTTTTTCACAAGGC-3’ and

5’-GGATTAATAGCTCCAAAAATCACCGC-3’ (for amplification of 3’ fragment).

The restriction sites are underlined. The plasmids to delete the genes were constructed

in a two step procedure as depicted in Figure 2.16. For deletion of pglB, the 5’

fragment of pglB was amplified by PCR using C. jejuni 81-176 genomic DNA as a

template. This fragment with a size of 1 kbp was cut with PvuI and cloned into

pILL600 (Labigne-Roussel et al., 1988) using the restriction sites ScaI and PvuI

upstream of the Kmr cassette. Next, the 3’ fragment was amplified by PCR using C.

jejuni 81-176 genomic DNA as a template. This fragment with a size of 1 kbp was cut

with NdeI and Eco31I and cloned into the resulting plasmid cleaved with VspI and

Eco31I. These restriction sites are located downstream of the Kmr cassette. For

deletion of acrA and ompH1, the 3’ fragments were amplified by PCR using C. jejuni

81-176 genomic DNA as a template, cut with VspI and cloned into pILL600 cleaved

with the same enzyme. Next, the 5’ fragments were amplified by PCR using C. jejuni

81-176 genomic DNA as a template, the fragments were cut with ScaI and PvuI and

72

cloned into the resulting plasmid cleaved with the same enzymes. These plasmids

containing aphA inserted in the same transcriptional orientation as the gene of interest

were used to transform C. jejuni 81-176 by natural transformation (Guerry et al.,

1994), and mutants were selected on MH agar containing kanamycin. All mutants

were characterized by PCR to confirm that the incoming plasmid DNA had integrated

by a double cross-over.

KmR

Eco31IScaI PvuI VspI

KmR

Eco31IScaI PvuI VspI

KmR

pILL600

Eco31IScaI PvuI VspI

KmR

pILL600

Eco31IScaI PvuI VspI

KmR

Eco31IScaI PvuI VspI

KmR

Eco31IScaI PvuI VspI

Figure 2.16: Two step cloning procedure to construct plasmids for deletion of pglB, acrA, andompH1 in C. jejuni. 5’ and 3’ fragments of the genes to be deleted with a size of 1 kbp were amplifiedby PCR. Restriction sites were introduced to clone the PCR fragments upstream and downstream of theKmr cassette of the plasmid pILL600. The direction of transcription of the C. coli aphA gene is markedwith an arrow. The product of the aphA gene confers resistance to kanamycin.

2.4.6 Construction of plasmids containing the pgl locus with

mutations in the pglB locus

To introduce the mutations into the pglB locus, the locus was subcloned (Figure 2.17).

A 4.5 kbp MluI-XbaI fragment from pBTLPS (Fry et al., 1998) was cloned into

pET16b cleaved by the same enzymes. A 1.1 kbp BglII-XbaI fragment from the

73

resulting plasmid containing the 3'-end of the pglB gene was cloned into the pSP73

cleaved by BglII and XbaI. The resulting plasmid was used as a template to

mutagenise the pglB locus using Quik-Change mutagenesis kit provided by

Stratagene (La Jolla, CA, USA). The following primers were used to generate the

different point mutations: 5’-

GATTATGTGGTAACTTGGGCGGCTTATGGTTATCCTGTGCG-3’ and 5’-

CGCACAGGATAACCATAAGCCGCCCAAGTTACCACATAATC-3’ (W458A,

D459A; Fnu4HI), 5'-

GTAACTTGGTGGGATGAAGAGTATCCTTGTGCGTTATTA-TAG-3’ and 5'-

CTATAATAACGCACAGGAT-ACTCTTCATCCCACCAAGTTAC -3' (Y460E

G461E; EarI), 5'-GTAACTTGGTGGGATGAGGGTTATCCTGTGCGTTAT-

TATAG-3' and 5'–

CTATAATAACGCACAGGATAACCCTCATCCCACCAAGTTA-3' (Y460E;

FokI), 5'–GTAACTTGGTGGGATTATGAGTACCCTGTGCGTTATTATAG-3' and

5'–CTATAATAACGCACAGGGTACTCATAATCCCACCAAGTTAC-3’ (G461E;

RsaI). The point mutations are underlined and the amino acid exchanges and the new

restriction sites given in brackets. Plasmids were sequenced by PCR-cycle sequencing

kit using the ABI Prism Dye Terminator cycle Sequencing Kit (Perkin Elmer, Foster

City, CA, USA) and sequencing analysis was performed with an ABI Prism 310

genetic Analyzer (Perkin Elmer, Foster City, CA, USA). The presence of the point

mutations was confirmed and the fragments were cloned back into the original

pBTLPS.

74

pBTLPS

MluI XbaI

pET16 b-4.5 kbp

pSP73-1.1kbp

BglII XbaI

BglII XbaIMluI

pBTLPS*

BglII XbaI

pSP73-1.1kbp*

BglII XbaI

MluI

MUTAGENESIS

MluI XbaI

pET16 b-4.5 kbp*

galE wlaB pglH pglI pglJ pglB pglA pglC pglD pglE pglF pglGgalE wlaB pglH pglI pglJ pglB pglA pglC pglD pglE pglF pglG

MluI XbaI BglIIBglIIBglII NcoI

SalI*

*

*

Figure 2.17: Construction of plasmids containing the pgl locus with point mutations in the pglBgene. The MluI-XbaI fragment from pBTLPS was cloned into pET16b cleaved by the same enzymes.From the resulting plasmid, a BglII-XbaI fragment was cloned into pSP73 cleaved by the sameenzymes. This plasmid was used as a template to introduce point mutations with Quick-Changemutagenesis kit. After introduction of the point mutations, the BglII-XbaI fragment was cloned backinto pBTLPS by two ligations. The point mutations are marked with an asterisk. The genes of the pgllocus are depicted with the restriction sites used. SalI and NcoI were used to construct the shuttleplasmid as described below.

2.4.7 Conjugation of C. jejuni

To complement the C. jejuni mutant cells, shuttle plasmids were mobilized from E.

coli DH5α containing pRK211.1 into C. jejuni mutant cells (Guerry et al., 1994).

Transconjugants were selected on MH agar containing kanamycin and

chloramphenicol.

To complement the pglB mutant, the following shuttle plasmid was constructed: a 16

kbp SalI–NcoI fragment of pBTLPS (see Figure 2.17) was cloned into the

chloramphenicol-resistant Campylobacter shuttle plasmid pRY111 (Yao et al., 1993)

digested with SalI and EcoRV. The resulting plasmid (pRY111(pgl)) contains the

75

complete pgl locus except a 186 bp deletion resulting in the truncation of 62 amino

acids at the C-terminus of PglG.

To complement acrA and ompH1 mutant cells, the following shuttle plasmids were

constructed: a 3.2 kbp EcoRV and a 2 kbp PstI-BglII fragment from 2 clones of a λ

ZAP Express (Stratagene, La Jolla, CA, USA) library constructed with Sau3A

partially digested C. jejuni 81-176 genomic DNA (Guerry et al., 2002), which

contained acrA and ompH1 respectively, were cloned into the shuttle plasmid

pRY111 cleaved by EcoRV and PstI and BamHI, respectively.

Membrane proteins from mutants were prepared as below except that cells (50 OD)

were collected from plates and resuspended in 50 mM Tris-HCl (pH 8.5) buffer

containing 10 mM NaCl, 1 mM DTT, 1 mM PMSF and DNase I.

2.4.8 Construction of plasmids encoding PglB-PhoA fusions

The following primers were used for the amplification of different fusion constructs

with pBTLPS serving as a template: 5'-

GGGATATCTTAGGATAAAAGATGTTGAAAAA-GAG-3' as the upstream primer

where the start codon is marked in bold letters and 5 different downstream primers:

5’-GGCCCGGATCCCCTGCTATCA-TATCTCTTGCGCC-3’ (1, 54.5 kDa), 5’-

GGCCCGGATCCCCATTAAAATACATAA-AGCC-3’ (2, 82.5 kDa) 5’-

GGCCCGGATCCCCTCTAAGCCCCCCTTTAAGGC-3’ (3, 89.6 kDa), 5’-

GCCCGGATCCCCCACATAATCTTCTATTGGC-3’ (4, 98.8 kDa), 5'-

CCGGATCCCCTTTCAACCCGCCAATTTTAAGTTTAAAAAACC-3' (5, 128.9

kDa) where the original stop codon is replaced by an codon for an amino acid (bold).

For the amplification of the phoA part, pCH2 (Hoffman and Wright 1985) served as a

template in PCR amplification and the following primers were used: 5'-

CCGGATCCGGGCATGCCT-GCAGCTCAGGGCG-3', 5'-GGTCGCGAAAAAAA-

CCACCCGGCAGCG-3'. The introduced restriction sites are underlined.

A two step cloning procedure was used to construct the plasmids encoding the PglB-

PhoA fusion constructs (Figure 2.18). PhoA was amplified by PCR and cloned into

the BamHI/NruI site of pACYC184. The five different pglB fusion constructs were

amplified and cloned in frame to phoA into the EcoRV/BamHI site of the generated

plasmid. The fusion constructs were expressed under the control of the constitutive

tetracycline promoter. The five different fusion constructs were transformed by

electroporation into E. coli CC118 cells.

76

pACYC184

BamHIEcoRV

tetR

NruI

pACYC184

BamHIEcoRV

tetR

NruI

BamHIEcoRV

tetR

NruIBamHIBamHIEcoRV

tetR

NruINruI

BamHIEcoRV

tetR

NruINruI

phoA

phoA

pglB

Figure 2.18: Two step cloning procedure to produce plasmids encoding PglB-PhoA fusionconstructs. PhoA was amplified by PCR. BamHI and NruI restriction sites were generated to clone thefragment into pACYC184 cleaved by the same enzymes. Truncated pglB with different sizes wasamplified by PCR and cloned into the plasmid containing phoA cleaved by EcoRV and BamHI.

2.4.9 Alkaline phosphatase activity assay

Exactly 0.2 OD600 from a late exponential phase growing cell culture were pelleted.

After resuspending in 1 ml 1 M Tris-Cl (pH 8.0), 1 drop of 0.1 % SDS and 3 drops of

CHCl3 were added and the solution was mixed. The samples were vortexed and kept

at 28 °C for 5 minutes. The reaction was started by addition of p-nitrophenyl-

phosphate (pNPP, 0.4 mg/ml) and stopped at the point where the solution turned into

a bright yellow (for active fractions) or after a period of approx. 70-80 minutes (for

77

inactive fractions) with 0.1 M potassium-dihydrophosphate (KH2PO4). The activity

was calculated in Miller units:

MU = (1000 * OD420)/(OD600 * time (min) * Volume (ml))

2.4.10 Purification of C. jejuni glycoproteins and identification by

mass spectrometry

C. jejuni cells were grown under microaerobic conditions at 42 °C for 18 h in liquid

cultures. The cells (1100 OD) were harvested, washed once with 0.9% NaCl solution

and resuspended in 40 ml 50 mM Tris-HCl (pH 8.5) buffer containing 1 M NaCl, 1

mM DTT, 1 mM PMSF and DNase I. The cells were broken by French Press and cell

debris removed by centrifugation for 10 min at 3,000 x g. The supernatant was

centrifuged at 200,000 x g for 60 min and membrane proteins were solubilized in 20

ml 50 mM Tris-HCl buffer (pH 8.5) containing 10 mM NaCl, 1 mM DTT, 1 mM

PMSF and 0.4% n-dodecyl-β-D-maltoside. Insoluble material was removed by

centrifugation (200,000 x g for 1 h). 40 mg solubilized membrane proteins (2 mg/ml)

were applied to a DEAE-Sepharose column (120 ml) equilibrated with solution A (50

mM Tris-HCl (pH 8.5), 10 mM NaCl, 1 mM DTT, 0.08% n-dodecyl-β-D-maltoside).

The proteins were eluted first with solution A for 250 min (1 ml/min) and then with a

linear gradient to 100 % solution B (50 mM Tris-HCl (pH 8.5), 1 M NaCl, 1 mM

DTT, 0.08 % n-dodecyl-β-D-maltoside) over 250 min. The glycoproteins of interest

eluted at 60 min and the fractions were dialyzed overnight at 4 °C against 50 mM

Tris-HCl (pH 8.5). This fraction was applied at 4 °C to a soybean agglutinin (SBA)

agarose column (Sigma-Aldrich, Buchs, CH, 4 ml) equilibrated with solution A. The

column was washed with 45 ml solution A and glycoproteins were eluted with

solution C (same as A plus 0.5 M galactose) at 45 °C.

The glycoproteins were separated by SDS-PAGE and stained by Gelcode, a

Coomassie-based stain (Gelcode, Pierce, Rockford, IL, USA). Proteins of 47 kDa and

35 kDa were cut from the gel using a scalpel, reduced with 10 mM DTT, the free

cysteine thiols alkylated with 50 mM iodoacetamide and trypsin-digested (sequencing

grade, Promega, Wallisellen, CH). The peptide fragments were analyzed by matrix-

assisted laser desorption/ionization (MALDI) mass spectrometry (Protein service

laboratory, ETH Zurich). The MALDI analyses were carried out on an Applied

Biosystems Voyager DE Elite MALDI mass spectrometer. The matrix used was 2,5-

78

dihydroxybenzoic acid. Proteins were identified by mass fingerprinting and database

searching, using the TrEMBL-database on ExPasy.

2.4.11 Expression of AcrA and the pgl locus in E. coli

AcrA of C. jejuni was amplified by PCR using C. jejuni 81-176 genomic DNA with

the following primers: 5’-GGAATTCCATATGAAATTATTTCAAAAAAATACT-

ATTTTGC-3’ and 5’-

CCGCTCGAGTTGTGCTCCAATTTCTTTAACTTCGCTACC-3’ where the

restriction sites are underlined. The XhoI site replaces the original stop codon of the

acrA gene. The PCR product was cloned into NdeI and XhoI restriction sites of

pET24b. This led to the expression of AcrA with a hexahis-tag at the C-terminus. To

express the pgl locus in E. coli, the 16 kbp SalI–BamHI fragment containing the pgl

gene cluster was cut from the shuttle plasmid pRY111(pgl) and cloned into

pACYC184 cleaved by SalI and BamHI. The resulting plasmid contains the complete

pgl locus except to 186 bp deletion resulting in the truncation of 62 amino acids at the

C-terminus of PglG. The different plasmids were electroporated into E. coli

BL21(DE3) cells and transformants were selected on LB agar containing kanamycin

and chloramphenicol. Expression of AcrA was induced with 1 mM IPTG. Membrane

proteins of these cells were prepared as described for C. jejuni cells above.

79

3 References

Aebi, M., Gassenhuber, J., Domdey, H. and te Heesen, S. (1996). Cloning and

characterization of the ALG3 gene of Saccharomyces cerevisiae. Glycobiology

6: 439-444.

Akiyama, Y. and Ito, K. (1993). Folding and assembly of bacterial alkaline

phosphatase in vitro and in vivo. J Biol Chem 268: 8146-8150.

Alm, R. A., Guerry, P., Power, M. E. and Trust, T. J. (1992). Variation in antigenicity

and molecular weight of Campylobacter coli VC167 flagellin in different

genetic backgrounds. J Bacteriol 174: 4230-4238.

Altmann, F., Staudacher, E., Wilson, I. B. H. and März, L. (1999). Insect cells as

hosts for the expression of recombinant glycoproteins. Glycoconj J 16: 109-

123.

Avila-Sakar, A. J., Misaghi, S., Wilson-Kubalek, E. M., Downing, K. H., Zgurskaya,

H., Nikaido, H. and Nogales, E. (2001). Lipid-layer crystallization and

preliminary three-dimensional structural analysis of AcrA, the periplasmic

component of a bacterial multidrug efflux pump. J Struct Biol 136: 81-88.

Bättig, P. (1998). Characterization of high-copy number suppressors of ∆ost4 in

Saccharmoyces cerevisiae. Institute of Microbiology. Zuerich, ETH Zuerich.

Bause, E., Breuer, W. and Peters, S. (1995). Investigation of the active site of

oligosaccharyltransferase from pig liver using synthetic tripeptides as tools.

Biochem J 312: 979-985.

Bause, E. and Legler, G. (1981). The role of the hydroxy amino acid in the triplet

sequence Asn-Xaa-Thr(Ser) for the N-glycosylation step during glycoprotein

biosynthesis. Biochem J 195: 639-644.

Bause, E., Wesemann, M., Bartoschek, A. and Breuer, W. (1997). Epoxyethylglycyl

peptides as inhibitors of oligosaccharyltransferase: double-labelling of the

active site. Biochem J 322: 95-102.

Black, R. E., Levine, M. M., Clements, M. L., Hughes, T. P. and Blaser, M. J. (1988).

Experimental Campylobacter jejuni infection in humans. J Infect Dis 157:

472-479.

80

Breuer, W. and Bause, E. (1995). Oligosaccharyl transferase is a constitutive

component of an oligomeric protein complex from pig liver endoplasmic

reticulum. Eur J Biochem 228: 689-696.

Breuer, W., Klein, R. A., Hardt, B., Bartoschek, A. and Bause, E. (2001).

Oligosaccharyltransferase is highly specific for the hydroxy amino acid in

Asn-Xaa-Thr/Ser. FEBS Lett 501: 106-110.

Bugg, T. D. H. and Brandish, P. E. (1994). From peptidoglycan to glycoproteins:

common features of lipid-linked oligosaccharide biosynthesis. FEMS

Microbiol Lett 119: 255-262.

Burda, P. and Aebi, M. (1998). The ALG10 locus of Saccharomyces cerevisiae

encodes the α-1,2 glucosyltransferase of the endoplasmic reticulum: the

terminal glucose of the lipid-linked oligosaccharide is required for efficient N-

linked glycosylation. Glycobiology 8: 455-462.

Burda, P. and Aebi, M. (1999). The dolichol pathway of N-linked glycosylation.

Biochim Biophys Acta 1426: 239-257.

Burda, P., Jakob, C. A., Beinhauer, J., Hegemann, J. H. and Aebi, M. (1999). Ordered

assembly of the asymmetrically branched lipid-linked oligosaccharide in the

endoplasmic reticulum is ensured by the substrate specificity of the individual

glycosyltransferases. Glycobiology 9: 617-625.

Burda, P., te Heesen, S. and Aebi, M. (1996). Stepwise assembly of the lipid-linked

oligosaccharide in the endoplasmic reticulum of Saccharomyces cerevisiae:

identification of the ALG9 gene encoding a putative mannosyl transferase.

Proc Natl Acad Sci USA 93: 7160-7165.

Chang, C. N., Kuang, W. J. and Chen, E. Y. (1986). Nucleotide sequence of the

alkaline phosphatase gene of Escherichia coli. Gene 44: 121-125.

Chang, G. and Roth, C. B. (2001). Structure of MsbA from E. coli: a homolog of the

multidrug resistance ATP binding cassette (ABC) transporters. Science 293:

1793-1800.

Chi, J. H., Roos, J. and Dean, N. (1996). The OST4 gene of Saccharomyces cerevisiae

encodes an unusually small protein required for normal levels of

oligosaccharyltransferase activity. J Biol Chem 271: 3132-3140.

81

Das, R. C. and Heath, E. C. (1980). Dolichyldiphosphoryloligosaccharide-protein

oligosaccharyltransferase; Solubilization, purification, and properties. Proc

Natl Acad Sci USA 77: 3811-3815.

Dekker, P. J. T., Müller, H., Rassow, J. and Pfanner, N. (1996). Characterization of

the preprotein translocase of the outer mitochondrial membrane by blue native

electrophoresis. Biol Chem 377: 535-538.

Dietzler, D. N. and Strominger, J. L. (1973). Characterization of the 4-Acetamido-4,6-

dideoxyhexoses from Escherichia coli strains. J Biol Chem 248: 104-109.

Dobos, K. M., Khoo, K. H., Swiderek, K. M., Brennan, P. J. and Belisle, J. T. (1996).

Definition of the full extent of glycosylation of the 45-kilodalton glycoprotein

of Mycobacterium tuberculosis. J Bacteriol 178: 2498-506.

Doig, P., Kinsella, N., Guerry, P. and Trust, T. J. (1996). Characterization of a post-

translational modification of Campylobacter flagellin: identification of a sero-

specific glycosyl moiety. Mol Microbiol 19: 379-387.

Erickson, P. R. and Herzberg, M. C. (1993). Evidence for the covalent linkage of

carbohydrate polymers to a glycoprotein from Streptococcus sanguis. J Biol

Chem 268: 23780-23783.

Figurski, D. H. and Helinski, D. R. (1979). Replication of an origin-containing

derivative of plasmid RK2 dependent on a plasmid function provided in trans.

Proc Natl Acad Sci USA 76: 1648-1652.

Fry, B. N., Feng, S., Chen, Y. Y., Newell, D. G., Coloe, P. J. and Korolik, V. (2000).

The galE gene of Campylobacter jejuni is involved in lipopolysaccharide

synthesis and virulence. Infect Immun 68: 2594-2601.

Fry, B. N., Korolik, V., ten Brinke, J. A., Pennings, M. T. T., Zalm, R., Teunis, B. J.

J., Coloe, P. J. and van der Zeijst, B. A. M. (1998). The lipopolysaccharide

biosynthesis locus of Campylobacter jejuni 81116. Microbiology 144: 2049-

2061.

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.

Grabenhorst, E., Schlenke, P., Pohl, S., Nimtz, M. and Conradt, H. S. (1999). Genetic

engineering of recombinant glycoproteins and the glycosylation pathway in

mammalian host cells. Glycoconj J 16: 81-97.

82

Guerry, P., Doig, P., Alm, R. A., Burr, D. H., Kinsella, N. and Trust, T. J. (1996).

Identification and characterization of genes required for post-translational

modification of Campylobacter coli VC167 flagellin. Mol Microbiol 19: 369-

378.

Guerry, P., Szymanski, C. M., Prendergast, M. M., Hickey, T. E., Ewing, C. P.,

Pattarini, D. L. and Moran, A. P. (2002). Phase variation of Campylobacter

jejuni 81-176 lipooligosaccharide affects ganglioside mimicry and

invasiveness in vitro. Infect Immun 70: 787-793.

Guerry, P., Yao, R., Alm, R. A., Burr, D. H. and Trust, T. J. (1994). Systems of

experimental genetics for Campylobacter species. Methods Enzymol 235: 474-

481.

Guthrie, C. and Fink, G. R. (1991). Guide to yeast genetics and molecular biology.

Methods Enzymol 194: 3-37.

Helenius, A. and Aebi, M. (2001). Intracellular functions of N-linked glycans. Science

291: 2364-2369.

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.

Hoffman, C. S. and Wright, A. (1985). Fusions of secreted proteins to alkaline

phosphatase: an approach for studying protein secretion. Proc Natl Acad Sci

USA 82: 5107-5111.

Huffaker, T. C. and Robbins, P. W. (1983). Yeast mutants deficient in protein

glycosylation. Proc Natl Acad Sci USA 80: 7466-7470.

Imperiali, B., Shannon, K. L. and Rickert, K. W. (1992a). Role of peptide

conformation in asparagine-linked glycosylation. J Am Chem Soc 114: 7942-

7944.

Imperiali, B., Shannon, K. L., Unno, M. and Rickert, K. W. (1992b). A mechanistic

proposal for asparagine-linked glycosylation. J Am Chem Soc 114: 7944-7945.

Jackson, B. J., Kukuruzinska, M. A. and Robbins, P. (1993). Biosynthesis of

asparagine-linked oligosaccharides in Saccharomyces cerevisiae: the alg2

mutation. Glycobiology 3: 357-364.

Jakob, C. A., te Heesen, S., Aebi, M. and Roth, J. (1998). Genetically tailoring of N-

linked oligosaccharides: the role of glucose residues in glycoprotein

processing in Saccharomyces cerevisiae in vivo. Glycobiology 8: 155-164.

83

Jennings, M. P., Virji, M., Evans, D., V., F., Srikhanta, Y. N., Steeghs, L., van der

Ley, P. and Moxon, E. R. (1998). Identification of a novel gene involved in

pilin glycosylation in Neisseria meningitidis. Mol Microbiol 29: 975-984.

Karaoglu, D., Kelleher, D. J. and Gilmore, R. (1995). Functional characterization of

Ost3p. Loss of the 34-kDa subunit of the Saccharomyces cerevisiae

oligosaccharyltransferase results in biased underglycosylation of acceptor

substrates. J Cell Biol 130: 567-577.

Karaoglu, D., Kelleher, D. J. and Gilmore, R. (1997). The highly conserved Stt3

protein is a subunit of the yeast oligosaccharyltransferase and forms a

subcomplex with Ost3p and Ost4p. J Biol Chem 272: 32513-32520.

Karaoglu, D., Kelleher, D. J. and Gilmore, R. (2001). Allosteric regulation provides a

molecular mechanism for preferential utilization of the fully assembled

dolichol-linked oligosaccharide by the yeast oligosaccharyltransferase.

Biochemistry 40: 12193-12206.

Kawabe, T., Fujihira, E. and Yamaguchi, A. (2000). Molecular construction of a

multidrug exporter system, AcrAB: molecular interaction between AcrA and

AcrB, and cleavage of the N-terminal signal sequence of AcrA. J Biochem

128: 195-200.

Kelleher, D. J. and Gilmore, R. (1994). The Saccharomyces cervisiae

oligosaccharyltransferase is a protein complex composed of Wbp1p, Swp1p,

and four additional polypeptides. J Biol Chem 269: 12908-12917.

Kelleher, D. J., Kreibich, G. and Gilmore, R. (1992). Oligosaccharyltransferase

activity is associated with protein complex composed of ribophorins I and II

and a 48 kD protein. Cell 69: 55-65.

Ketley, J. M. (1997). Pathogenesis of enteric infection by Campylobacter.

Microbiology 143: 5-21.

Knauer, R. and Lehle, L. (1994). The N-oligosaccharyltransferase complex from

yeast. FEBS Lett 344: 83-86.

Knauer, R. and Lehle, L. (1999a). The Oligosaccharyltransferase complex from

Saccharomyces cerevisiae. Isolation of the OST6 gene, its synthetic interaction

with OST3, and analysis of the native complex. J Biol Chem 274: 17249-

17256.

Knauer, R. and Lehle, L. (1999b). The oligosaccharyltransferase complex from yeast.

Biochim Biophys Acta 1426: 259-273.

84

Korlath, J. A., Osterholm, M. T., Judy, L. A., Forfang, J. C. and Robinson, R. A.

(1985). A point-source outbreak of campylobacteriosis associated with

consumption of raw milk. J Infect Dis 152: 592-596.

Kornfeld, R. and Kornfeld, S. (1985). Assembly of asparagine-linked

oligosaccharides. Annu Rev Biochem 54: 631-664.

Labigne-Roussel, A., Courcoux, P. and Tompkins, L. (1988). Gene disruption and

replacement as a feasible approach for mutagenesis of Campylobacter jejuni. J

Bacteriol 170: 1704-1708.

Lechner, J. and Wieland, F. (1989). Structure and biosynthesis of prokaryotic

glycoproteins. Annu Rev Biochem 58: 173-194.

Linton, D., Allan, E., Karlyshev, A. V., Cronshaw, A. D. and Wren, B. W. (2002).

Identification of N-acetylgalactosamine-containing glycoproteins PEB3 and

CgpA in Campylobacter jejuni. Mol Microbiol 43: 497 - 508.

Linton, D., Karlyshev, A. V., Hitchen, P. G., Morris, H. R., Dell, A., Gregson, N. A.

and Wren, B. W. (2000). Multiple N-acetyl neuraminic acid synthetase (neuB)

genes in Campylobacter jejuni: identification and characterization of the gene

involved in sialylation of lipo-oligosaccharide. Mol Microbiol 35: 1120-1134.

Ma, D., Cook, D. N., Alberti, M., Pon, N. G., Nikaido, H. and Hearst, J. E. (1993).

Molecular cloning and characterization of acrA and acrE genes of Escherichia

coli. J Bacteriol 175: 6299-6313.

Malissard, M., Zeng, S. and Berger, E. G. (1999). The yeast expression system for

recombinant glycosyltransferases. Glycoconj J 16: 125-139.

Matsuhashi, M. and Strominger, J. L. (1966). Thymidine diphosphate 4-acetamido-

2,6-dideoxyhexoses. III. Purification and properties of thymidine diphosphate

4-keto-6-deoxy-D-glucose transaminase from Escherichia coli strain B. J Biol

Chem 241: 4738-4744.

Meinersmann, R. J., Hiett, K. L. and Tarplay, A. (1997). Cloning of an outer

membrane protein gene from Campylobacter jejuni. Curr Microbiol 34: 360-

366.

Messner, P. (1997). Bacterial glycoproteins. Glycoconj J 14: 3-11.

Moens, S. and Vanderleyden, J. (1997). Glycoproteins in prokaryotes. Arch Microbiol

168: 169-175.

Nachamkin, I., Allos, B. M. and Ho, T. (1998). Campylobacter species and Guillain-

Barré syndrome. Clin Microbiol Rev 11: 555-567.

85

Newell, D. G. (2001). Animal models of Campylobacter jejuni colonization and

disease and the lessons to be learned from similar Helicobacter pylori models.

J Appl Microbiol: 57S-67S.

Nikaido, H. (2000). How do exported proteins and antibiotics bypass the periplasm in

Gram-negative bacterial cells? Trends Microbiol 8: 481-483.

Nikaido, H. and Zgurskaya, H. I. (2001). AcrAB and related multidrug efflux pumps

of Escherichia coli. J Mol Microbiol Biotechnol 3: 215-218.

Parkhill, J., Wren, B. W., Mungall, K., Ketley, J. M., Churcher, C., Basham, D.,

Chillingworth, T., Davies, R. M., Feltwell, T., Holroyd, S., Jagels, K.,

Karlyshev, A. V., Moule, S., Pallen, M. J., Penn, C. W., Quail, M. A.,

Rajandream, M. A., Rutherford, K. M., van Vliet, A. H. M., Whitehead, S. and

Barrell, B. G. (2000). The genome sequence of the food-borne pathogen

Campylobacter jejuni reveals hypervariable sequences. Nature 403: 665-668.

Pathak, R., Hendrickson, T. L. and Imperiali, B. (1995). Sulfhydryl modification of

the yeast Wbp1p inhibits oligosaccharyl transferase activity. Biochemistry 34:

4179-4185.

Pathak, R. and Imperiali, B. (1997). A dual affinity tag on the 64-kDa Nlt1p subunit

allows the rapid characterization of mutant yeast oligosaccharyl transferase

complexes. Arch Biochem Biophys 338: 1-6.

Pawelec, D. P., Korsak, D., Wyszynska, A. K., Rozynek, E., Popowski, J. and

Jagusztyn-Krynicka, E. K. (2000). Genetic diversity of the Campylobacter

genes coding immunodominant proteins. FEMS Microbiol Lett 185: 43-49.

Pei, Z. H., Ellison, R. T. and Blaser, M. J. (1991). Identification, purification, and

characterization of major antigenic proteins of Campylobacter jejuni. J Biol

Chem 266: 16363-16369.

Power, P. M., Roddam, L. F., Dieckelmann, M., Srikhanta, Y. N., Tan, Y. C.,

Berrington, A. W. and Jennings, M. P. (2000). Genetic characterization of

pilin glycosylation in Neisseria meningitidis. Microbiology 146: 967-979.

Pumbwe, L. and Piddock, L. J. V. (2002). Identification and molecular

characterisation of CmeB, a Campylobacter jejuni multidrug efflux pump.

FEMS Microbiol Lett 206: 185-189.

Raetz, C. R. H. (1996). Bacterial lipopolysaccharides: a remarkable family of

bioactive macroamphiphiles. Escherichia coli and Salmonella: Cellular and

86

Molecular Biology. Neidhardt, F. C., Curtiss, R., Ingraham, J. L. et al.

Washington D.C., American Society for Microbiology: 1035 -1063.

Reeves, P. R., Hobbs, M., Valvano, M. A., Skurnik, M., Whitfield, C., Coplin, D.,

Kido, N., Klena, J., Maskell, D., Raetz, C. R. and Rick, P. D. (1996). Bacterial

polysaccharide synthesis and gene nomenclature. Trends Microbiol 4: 495-

503.

Reiss, G., te Heesen, S., Gilmore, R., Zufferey, R. and Aebi, M. (1997). A specific

screen for oligosaccharyltransferase mutations identifies the 9 kDa OST5

protein required for optimal activity in vivo and in vitro. EMBO J 16: 1164-

1172.

Reiss, G., te Heesen, S., Zimmerman, J., Robbins, P. W. and Aebi, M. (1996). The

ALG6 locus of Saccharomyces cervisiae required for glucosylation in the N-

linked glycosylation pathway. Glycobiology 6: 493-498.

Sailer, A. and Weissmann, C. (1991). A sensitive and rapid protein assay impervious

to detergents. Technique 3: 37-38.

Schäffer, C., Graninger, M. and Messner, P. (2001). Prokaryotic glycosylation.

Proteomics 1: 248-261.

Schäffer, C. and Messner, P. (2001). Glycobiology of surface layer proteins.

Biochimie 83: 591-599.

Schägger, H. (1995). Native electrophoresis for isolation of mitochondrial oxidative

phosphorylation protein complexes. Methods Enzymol 260: 190-202.

Schägger, H., Cramer, W. A. and von Jagow, G. (1994). Analysis of molecular

masses and oligomeric states of protein complexes by blue native

electrophoresis and isolation of membrane protein complexes by two-

dimensional native electrophoresis. Anal Biochem 217: 220-230.

Schägger, H. and von Jagow, G. (1991). Blue native electrophoresis for isolation of

membrane protein complexes in enzymatically active form. Anal Biochem

199: 223-231.

Sharma, C. B., Lehle, L. and Tanner, W. (1981). N-glycosylation of yeast proteins.

Characterization of the solubilized oligosaccharyl transferase. Eur J Biochem

116: 101-108.

Silberstein, S., Collins, P. G., Kelleher, D. J. and Gilmore, R. (1995a). The essential

OST2 gene encodes the 16-kDa subunit of the yeast oligosaccharyltransferase,

87

a highly conserved protein expressed in diverse eukaryotic organisms. J Cell

Biol 131: 371-383.

Silberstein, S., Collins, P. G., Kelleher, D. J., Rapiejko, P. J. and Gilmore, R. (1995b).

The alpha subunit of the Saccharomyces cerevisiae oligosaccharyltransferase

is essential for vegetative growth of yeast and is homologous to mammalian

ribophorin I. J Cell Biol 128: 525-536.

Silberstein, S. and Gilmore, R. (1996). Biochemistry, molecular biology, and genetics

of the oligosaccharyltransferase. FASEB J 10: 849-858.

Silberstein, S., Kelleher, D. J. and Gilmore, R. (1992). The 48-kDa subunit of the

mammalian oligosaccharyltransferase complex is homologous to the essential

yeast protein Wbp1. J Biol Chem 267: 23658-23663.

Spirig, U. (1999). The oligosaccharyltransferase complex of Saccharomyces

cerevisiae. Institute of Microbiology. Zuerich, ETH Zuerich.

Spirig, U., Glavas, M., Bodmer, D., Reiss, G., Burda, P., Lippuner, V., te Heesen, S.

and Aebi, M. (1997). The STT3 protein is a component of the yeast

oligosaccharyltransferase complex. Mol Gen Genet 256: 628-637.

Spiro, M. J., Spiro, R. G. and Bhoyroo, V. D. (1976). Lipid-saccharide intermediates

in glycoprotein biosynthesis. II. Studies on the structure of an oligosaccharide-

lipid from thyroid. J Biol Chem 251: 6409-6419.

Stagljar, I., te Heesen, S. and Aebi, M. (1994). New phenotype of mutations deficient

in the glycosylation of the lipid-linked oligosaccharide: cloning of the ALG8

locus. Proc Natl Acad Sci USA 91: 5977-5981.

Stimson, E., Virji, M., Makepeace, K., Dell, A., Morris, H. R., Payne, G., Saunders, J.

R., Jennings, M. P., Barker, S., Panico, M., Blench, I. and Moxon, E. R.

(1995). Meningococcal pilin: a glycoprotein substituted with digalactosyl 2,4-

diacetamido-2,4,6-trideoxyhexose. Mol Microbiol 17: 1201-1214.

Szymanski, C. M., Burr, D. H. and Guerry, P. (2002). Campylobacter protein

glycosylation affects host cell interactions. Infect Immun 70: 2242-2244.

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.

te Heesen, S., Janetzky, B., Lehle, L. and Aebi, M. (1992). The yeast WBP1 is

essential for oligosaccharyl transferase activity in vivo and in vitro. EMBO J

11: 2071-2075.

88

te Heesen, S., Knauer, R., Lehle, L. and Aebi, M. (1993). Yeast Wbp1p and Swp1p

form a protein complex essential for oligosaccharyl transferase activity.

EMBO J 12: 279-284.

te Heesen, S., Lehle, L., Weissmann, A. and Aebi, M. (1994). Isolation of the ALG5

locus, encoding the UDP-glucose:dolichyl-phosphate glucosyltransferase from

Saccharomyces cerevisiae. Eur J Biochem 224: 71-79.

Thibault, P., Logan, S. M., Kelly, J. F., Brisson, J. R., Ewing, C. P., Trust, T. J. and

Guerry, P. (2001). Identification of the carbohydrate moieties and

glycosylation motifs in Campylobacter jejuni flagellin. J Biol Chem 276:

34862-34870.

Trieu-Cuot, P., Gerbaud, G., Lambert, T. and Courvalin, P. (1985). In vivo transfer of

genetic information between gram-positive and gram-negative bacteria.

EMBO J 4: 3583-3587.

van Geest, M. and Lolkema, J. S. (2000). Membrane topology and insertion of

membrane proteins: search for topogenic signals. Microbiol Mol Biol Rev 64:

13-33.

van Vliet, A. H. M. and Ketley, J. M. (2001). Pathogenesis of enteric Campylobacter

infection. J Appl Microbiol 90: 45S-56S.

Vijayraghavan, U., Company, M. and Abelson, J. (1989). Isolation and

characterization of pre-mRNA splicing mutants of Saccharomyces cerevisiae.

Genes Dev 3: 1206-1216.

Whitfield, C. (1995). Biosynthesis of lipopolysaccharide O antigens. Trends

Microbiol 3: 178-185.

Whitfield, C. and Roberts, I. S. (1999). Structure, assembly and regulation of

expression of capsules in Escherichia coli. Mol Microbiol 31: 1307-1319.

Wieland, F. T., Gleason, M. L., Serafini, M. L. and Rothman, J. E. (1987). The rate of

bulk flow from the endoplasmic reticulum to the cell surface. Cell 50: 289-

300.

Wood, A. C., Oldfield, N. J., O'Dwyer, C. A. and Ketley, J. M. (1999). Cloning,

mutation and distribution of a putative lipopolysaccharide biosynthesis locus

in Campylobacter jejuni. Microbiology 145: 379-388.

Wooldridge, K. G. and Ketley, J. M. (1997). Campylobacter-host cell interactions.

Trends Microbiol 5: 96-102.

89

Wu, L. and Welker, N. E. (1991). Cloning and characterization of a glutamine

transport operon of Bacillus stearothermophilus NUB36: effect of temperature

on regulation of transcription. J Bacteriol 173: 4877-4888.

Yan, Q. and Lennarz, W. J. (1999). Oligosaccharyltransferase: a complex

multisubunit enzyme of the endoplasmic reticulum. Biochem Biophys Res

Commun 266: 684-689.

Yao, R., Alm, R. A., Trust, T. J. and Guerry, P. (1993). Construction of new

Campylobacter cloning vectors and a new mutational cat cassette. Gene 130:

127-130.

Zeitler, R., Hochmuth, E., Deutzmann, R. and Sumper, M. (1998). Exchange of Ser-4

for Val, Leu or Asn in the sequon Asn-Ala-Ser does not prevent N-

glycosylation of the cell surface glycoprotein from Halobacterium halobium.

Glycobiology 8: 1157-1164.

Zgurskaya, H. I. and Nikaido, H. (1999). AcrA is a highly asymmetric protein capable

of spanning the periplasm. J Mol Biol 285: 409-420.

Zgurskaya, H. I. and Nikaido, H. (2000). Multidrug resistance mechanisms: drug

efflux across two membranes. Mol Microbiol 37: 219-225.

Zhou, Z., White, K. A., Polissi, A., Georgopoulos, C. and Raetz, C. R. (1998).

Function of Escherichia coli MsbA, an essential ABC family transporter, in

lipid A and phospholipid biosynthesis. J Biol Chem 273: 12466-12475.

Zufferey, R., Knauer, R., Burda, P., Stagljar, I., te Heesen, S., Lehle, L. and Aebi, M.

(1995). STT3, a highly conserved protein required for yeast oligosaccharyl

transferase activity in vivo. EMBO J 14: 4949-4960.

90

Curriculum vitae

Name Michael Wacker

Date of birth 25th of April 1973,

Place of birth Geneva

Citizen of Beatenberg BE

Education

1998 – 2002 PhD thesis under the guidance of Prof. Dr. M.

Aebi at the Institute of Microbiology, Swiss

Federal Institute of Technology, Zurich

1993 – 1998 Studies of biochemistry at the Swiss Federal

Institute of Technology, Zurich

1986 – 1992 Secondary education (Gymnasium) in Zurich;

School leaving examination, Matura type B

1980 – 1986 Primary education in Pfaffhausen