View
214
Download
0
Category
Preview:
Citation preview
Charakterisierung und physiologische Relevanz des Membranlipids Phosphatidylcholin für das
pflanzenpathogene Bakterium Agrobacterium tumefaciens
Dissertation zur Erlangung des Grades eines Doktors der Naturwissenschaften
der Fakultät für Biologie und Biotechnologie an der Internationalen Graduiertenschule Biowissenschaften
der Ruhr-Universität Bochum
angefertigt im Lehrstuhl für Biologie der Mikroorganismen
vorgelegt von Sonja Klüsener
aus
Wuppertal
Bochum April, 2010
Characterization and physiological relevance of the membrane lipid phosphatidylcholine in the
phytopathogenic bacterium Agrobacterium tumefaciens
Dissertation zur Erlangung des Grades eines Doktors der Naturwissenschaften
der Fakultät für Biologie und Biotechnologie an der Internationalen Graduiertenschule Biowissenschaften
der Ruhr-Universität Bochum
angefertigt im Lehrstuhl für Biologie der Mikroorganismen
vorgelegt von Sonja Klüsener
aus
Wuppertal
Bochum April, 2010
Danksagung
Mein erster Dank gilt meinem Doktorvater Prof. Dr. Franz Narberhaus für die
Überlassung des Themas, die Bereitstellung des Arbeitsplatzes und die Betreuung.
Besonders möchte ich mich für die Förderung und Unterstützung im internationalen
wissenschaftlichen Austausch und bei der Veröffentlichung meiner Ergebnisse
bedanken.
Bei Herrn Prof. Dr. Matthias Rögner bedanke ich mich für die freundliche Übernahme
des Korreferats.
Ein ganz herzlicher Dank gilt Dr. Bernd Masepohl für zahlreiche konstruktive
Diskussionen sowie pünktliche und abwechslungsreiche Mittagspausen in der
Mensa. Ebenso danke ich Petra und Hanno für ihre fortwährende Hilfsbereitschaft
und Unterstützung in vielen kleinen und großen Dingen.
Ich danke Dr. Ehr-Min Lai und Dr. Yun-long Tsai (Academia Sinica, Taipei) für die
freundliche Betreuung in Taiwan und den produktiven wissenschaftlichen Austausch
während der letzten zwei Jahre. Dr. Kai Thormann (MPI, Marburg) danke ich für
seine Hilfsbereitschaft insbesondere bei der Untersuchung der Biofilme. Jun. Prof.
Dr. Julia Bandow danke ich für die Unterstützung bei den 2D-Gelanalysen, Knut
Büttner (Ernst-Moritz-Arndt-Universität, Greifswald) für die MALDI-MS-Analysen und
Ronald Gust (Freie Universität Berlin, Berlin) für die GFAAS-Analysen.
Mein Dank richtet sich auch an alle Mitarbeiter des Lehrstuhls für das gute
Arbeitsklima und die ständige Bereitschaft mit Rat und Tat zu helfen.
Den Bakterien-Pflanzen Mädels und Jan möchte ich für eine sehr gute
Zusammenarbeit und viele lustige Stunden im Labor/Büro danken. Besonders danke
ich Meriyem für ihre hilfreichen Ratschläge und das fleißige Korrekturlesen sowie
Christiane für die experimentelle Unterstützung.
Meiner Bachelorstudentin Kathinka und meinen Diplomanden Robbin und Philip
danke ich für ihr erfolgreiches Mitwirken an dieser Arbeit.
Ein herzlicher Dank gilt allen meinen Freunden für die kontinuierliche Unterstützung,
die grenzenlose Geduld, den Zuspruch und die vielen, vielen schönen Stunden
neben der Promotion. Dabei möchte ich ganz besonders Doro, Mirja, Ingo, Sina,
Julia, Alex, Nadja, Britta und Linda dafür danken, dass sie jederzeit für mich da
waren und mich unermüdlich bestärkt haben.
Mein ganz besonderer Dank geht an meine Familie für ihre fortwährende
Unterstützung bei jeder von mir getroffenen Entscheidung, Ihr Verständnis und das
Vertrauen in mich, wodurch sie einen wichtigen Beitrag zum Gelingen meiner
Doktorarbeit geleistet haben.
Inhaltsverzeichnis
I
I Inhaltsverzeichnis
I Inhaltsverzeichnis I
II Abkürzungen III
A Einleitung 1
1. Bedeutung von Phosphatidylcholin 1
1.1 Phosphatidylcholin in eukaryotischen Zellen 1
1.2 Phosphatidylcholin in prokaryotischen Zellen 4
2. Biosynthese von Phosphatidylcholin 8
2.1 Biosynthese von Phosphatidylcholin in Eukaryoten 8
2.2 Biosynthese von Phosphatidylcholin in Prokaryoten 11
3. Der Modellorganismus Agrobacterium tumefaciens 15
3.1 Der Infektionsprozess 15
3.2 Virulenzfaktoren 18
3.3 Biosynthese und Bedeutung von Phosphatidylcholin
in A. tumefaciens 21
4. Zielsetzung 23
B Expression and physiological relevance of Agrobacterium tumefaciens phosphatidylcholine biosynthesis genes 24
C Proteomic and transcriptomic characterization of a virulence-deficient phosphatidylcholine-negative Agrobacterium tumefaciens mutant 35
D Small heat-shock protein HspL is induced by VirB protein(s) and promotes VirB/D4-mediated DNA transfer in Agrobacterium tumefaciens 51
E Diskussion 63
1. Biosynthese von Phosphatidylcholin in A. tumefaciens 63
Inhaltsverzeichnis
II
2. Globale Auswirkungen der Phosphatidylcholin-Defizienz
in A. tumefaciens 67
3. Einfluss von Phosphatidylcholin auf die Biofilmbildung
und Motilität 71
4. Virulenzdefekt in Abwesenheit von Phosphatidylcholin
in A. tumefaciens 75
5. Einfluss von HspL auf das Typ IV-Sekretionssystem
in A. tumefaciens 80
F Zusammenfassung 85
G Summary 87
H Referenzen 89
I Publikationen 104
1. Originalartikel 104
2. Konferenzbeiträge 104
J Anhang 106
1. Curriculum vitae 106
2. Erklärung 107
3. Eigenanteil an den Publikationen 108
Abkürzungen
III
II Abkürzungen
Abb. Abbildung
ABC engl.: ATP-binding cassette
ADP Adenosindiphosphat
ATP Adenosintriphosphat
bp Basenpaare
bzw. beziehungsweise
ca. lat.: circa
CCT CTP:Phosphocholin-Cytidylyltransferase
CDP Cytidindiphosphat
CKI Cholin-Kinase
CL Cardiolipin
CMP Cytidinmonophosphat
CPT 1,2-Diacylglycerin-Cholin-Phosphotransferase
CTP Cytidintriphosphat
DAG Diacylglycerin
d.h. das heißt
DNA engl.: deoxyribonucleic acid
DMPE Dimethylphosphatidylethanolamin
Hcp engl.: hemolysin coregulated protein
Hsp Hitzeschockprotein
MMPE Monomethylphosphatidylethanolamin
PC Phosphatidylcholin
Pcs Phosphatidylcholinsynthase
PE Phosphatidylethanolamin
PG Phosphatidylglycerol
PI Phosphatidylinositol
PL Phospholipase
Pmt Phospholipid N-Methyltransferase
PS Phosphatidylserin
Pss Phosphatidylserinsynthase
RNA engl.: ribonucleic acid
ROSE engl.: repression of heat shock gene expression
Abkürzungen
IV
RT-PCR engl.: reverse transcription – polymerase chain reaction
SAH S-Adenosylhomocystein
SAM S-Adenosylmethionin
SDS engl.: sodium dodecylsulfate
sog. so genannt
Tab. Tabelle
Ti-Plasmid engl.: tumor-inducing plasmid
TIVSS Typ IV-Sekretionssystem
u.a. unter anderem
vir engl.: virulence
VLDL engl.: very low-density lipoprotein
w/v engl.: weight per volume
z.B. zum Beispiel
16SrRNA ribosomale RNA der Größe 16 S (Svedberg)
Einleitung
1
A Einleitung
1. Bedeutung von Phosphatidylcholin
Phosphatidylcholine (Lecithin; PC) sind weit verbreitete Bestandteile von
biologischen Membranen und gehören in die Gruppe der Phosphoglyceride. Diese
bestehen aus einem Glycerinrückgrat, dass an den sn1- und sn2-Positionen mit je
einer langkettigen Fettsäure verestert ist. An der dritten, endständigen
Hydroxylgruppe ist eine Phosphatgruppe gebunden. Diese bildet einen
Phosphorsäurediester, einerseits mit dem Glycerin und andererseits, im Fall von PC,
mit dem einwertigen Alkohol Cholin (Abb. 1A). Cholin trägt als quartäre
Ammoniumverbindung eine positive Ladung, während die Phosphatgruppe als Anion
vorliegt, so dass PC in die Gruppe der neutralen Phospholipide eingeordnet wird. PC
ist aufgrund des amphipatischen Charakters eine wichtige Komponente bei der
Ausbildung von Lipiddoppelschichten (Abb. 1B) und verleiht der Membran Flexibilität
und Anpassungsfähigkeit.
A)
B)
Abb. 1: Struktur von Phosphatidylcholin. A) PC setzt sich aus einem apolaren Teil (gelb), der die gebundenen Fettsäuren trägt und einer polaren Kopfgruppe (blau) zusammen. Die veresterten Fettsäuren variieren von Molekül zu Molekül. Die in diesem Beispiel gezeigten Fettsäuren sind die ungesättigte Ölsäure (oben) und die Palmitinsäure (unten). B) Schematische Darstellung einer Lipiddoppelschicht.
1.1 Phosphatidylcholin in eukaryotischen Zellen
Die Zusammensetzung eukaryotischer Membranen ist heterogen und variiert in
verschiedenen Organellen. Sogar innerhalb der Lipiddoppelschicht einer Membran
liegt eine asymmetrische Verteilung der Lipide vor (Zachowski, 1993). Der
exoplasmatische Teil der Plasmamembran besteht hauptsächlich aus PC,
Sphingomyelin und Glykosphingolipiden, während im cytoplasmatischen Teil
Einleitung
2
Phosphatidylserin (PS), Phosphatidylethanolamin (PE), Phosphatidsäure sowie
Phosphatidylinositol (PI) überwiegen (Yamaji-Hasegawa & Tsujimoto, 2006). Das
Vorkommen der Lipide ist in vielen Fällen an physiologische Prozesse angepasst.
Sphingomyeline können zusammen mit den Cholesterinen und Glykosphingolipiden
Mikrodomänen (lipid rafts) bilden, die in viele zelluläre Funktionen, z.B. der Endo-
und Exocytose, involviert sind (Rajendran & Simons, 2005). PS wird als Kofaktor
verschiedener Enzyme (z.B. Proteinkinase C und Raf-1 Kinase) beschrieben und
spielt darüber hinaus eine zentrale Rolle in pathologischen Prozessen (Vance &
Steenbergen, 2005).
Vorkommen und Funktionen von Phosphatidylcholin PC kommt als strukturgebende Komponente in allen eukaryotischen Membranen vor
und ist somit ein essentieller Bestandteil (Kent, 1990). In den Membranen von
Säugetieren ist PC mit mehr als 50 % das Hauptlipid, wobei in der Leber, im Gehirn,
im Herzen, in der Lunge sowie im Muskelgewebe die höchsten Konzentrationen
vorliegen (Zeisel, 1993). Neben dem Erhalt der Integrität von Zellmembranen ist PC
auch von wichtiger, physiologischer Bedeutung. PC ist eine essentielle Komponente
für die Sekretion des Lipoproteins VLDL (very low-density lipoprotein), welches für
den Export der in der Leber synthetisierten Triglyceride verantwortlich ist (Yao &
Vance, 1988; Yao & Vance, 1989). Des Weiteren ist PC in die Cholesterin-
Homöostase involviert. Mit einem PC-Anteil von 90 % der Gesamtphospholipide in
der Galle von Ratten ist es für den Cholesterin-Transport von zentraler Bedeutung
(LeBlanc et al., 1998). Außerdem ist PC mit einem Anteil von 60-80 % ein
Hauptbestandteil der Surfactantschicht der Lungenalveoli und für die
Aufrechterhaltung der Pulmonalaktivität wichtig (Kent, 1990; McMahon & Farrell,
1986).
Hydrolyse von Phosphatidylcholin Neben den direkten Funktionen von PC wird die PC-Hydrolyse durch Freisetzung
verschiedener sekundärer Botenstoffe als weit verbreiteter Signaltransduktionsweg
diskutiert (Billah & Anthes, 1990; Exton, 1994). Die PC-Hydrolyse durch die
Phospholipasen (PL) des Typs A2, C und D wird durch viele Wachstumsfaktoren,
Cytokine, Neurotransmitter und Hormone stimuliert (Exton, 1990; Shukla & Halenda,
1991). Die PLA2 katalysieren die Abspaltung der Acylgruppe an der sn2-Position von
Einleitung
3
PC durch Hydrolyse der Esterbindung (Clark et al., 1990), so dass die freigesetzte
Fettsäure und Lyso-PC als Produkte entstehen (Abb. 2). Die Hauptfunktion der PLA2-
Aktivität ist die Bereitstellung von Arachidonsäure, welche abhängig vom Zelltyp
weiter zu verschiedenen Eikosanoiden metabolisiert wird (Clark et al., 1991; Diez &
Mong, 1990). Darüber hinaus gibt es Hinweise, dass Arachidonsäure verschiedene
Isoformen der Proteinkinase C (α, β, γ, ε und ζ), die PC hydrolisierende PLD und die
Phosphatidsäure-Phosphohydrolase aktiviert (Murakami & Routtenberg, 1985;
Sekiguchi et al., 1987; Siddiqui & Exton, 1992).
PL des Typs C spalten PC zu Diacylglycerin (DAG) und Phosphocholin, während die
Produkte der PLD-Aktivität Phosphatidsäure und Cholin sind (Abb. 2) (Exton, 1994).
Im Gehirn von Säugetieren ist die PLD katalysierte PC-Hydrolyse mit der
Acetylcholin-Synthese gekoppelt und durch den produzierten Neurotransmitter bei
der Reizübertragung im Nervensystem von signifikanter Bedeutung (Blusztajn et al.,
1987). Neben der beschriebenen PC-Hydrolyse katalysiert die PLD eine weitere
Reaktion, die Transphosphatidylierung. Bei dieser Reaktion wird auch Cholin
freigesetzt, jedoch überträgt das Enzym eine Alkoholgruppe auf das Glycerin und
bildet damit den entsprechenden Phosphatidylalkohol (Exton, 1994).
Abb. 2: Angriffspunkte verschie-dener Phospholipasen. PC wird durch die Phospholipasen (PL) des Typs A2, C und D hydrolysiert. Dabei erfolgt die Spaltung an unterschiedlichen Stellen (durch Pfeile markiert), so dass die entstehenden Produkte variieren.
Der induzierte PC-Abbau und die damit verbundene Freisetzung von sekundären
Botenstoffen durch PLC’s und PLD’s spielt eine zentrale Rolle in der
Signaltransduktion (Billah & Anthes, 1990). Das primäre Produkt der PLD-Hydrolyse
Phosphatidsäure kann durch Phosphatidsäure-Phosphohydrolasen in DAG, durch
PLA2’s in Lysophosphatidsäure oder in CDP-Diacylglycerin für die Neusynthese von
Phospholipiden umgesetzt werden (Exton, 1994). Obwohl Phosphatidsäure und
Lysophosphatidsäure beide als Signalmoleküle fungieren, zeigen sie jedoch
unterschiedliche Effekte auf die Zellen. Phosphatidsäure hat prinzipiell Auswirkungen
auf intrazelluläre Proteine. Das Lipid ist in die Hyperoxid- (O2-) Produktion in
humanen Neutrophilen involviert (Bauldry et al., 1992) und führt in Fibroblasten zu
morphologischen Veränderungen, welche durch Aktinpolymerisation hervorgerufen
Einleitung
4
werden (Ha & Exton, 1993). Die Lysophosphatidsäure agiert als
extracytoplasmatisches Signal und vermittelt über eine Rezeptor-Interaktion PI-
Hydrolyse, Ca2+-Anstieg, Freisetzung von Arachidonsäure, Inhibierung der
Adenylatcyclase, PLD-Aktivierung und Chemotaxis (Durieux & Lynch, 1993; Exton,
1994; van der Bend et al., 1992). Im Gegensatz zur Phosphatidsäure, induziert
Lysophosphatidsäure eine Rho-abhängige Aktinpolymerisation (Ridley & Hall, 1992).
1.2 Phosphatidylcholin in prokaryotischen Zellen
Die Membranen der Modellorganismen Escherichia coli und Bacillus subtilis
enthalten als Hauptlipide Phosphatidylglycerol (PG), PE und Cardiolipin (CL) (Ames,
1968; op den Kamp et al., 1969). Es wurde angenommen, dass alle bakteriellen
Membranen eine ähnliche Zusammensetzung der Lipide aufweisen. Tatsächlich
existieren jedoch eine hohe Diversität und eine unterschiedliche Gewichtung der
membranbildenden Lipide in Prokaryoten. So wurden bakterielle Sphingolipide
identifiziert (Kunsman, 1973) sowie PS, Phosphatidsäure, PI oder die methylierten
PE-Derivate Monomethylphosphatidylethanolamin (MMPE), Dimethylphosphatidyl-
ethanolamin (DMPE) und PC, die bislang als typisch eukaryotische
Membrankomponenten galten (Johnson et al., 1970; Tornabene, 1973). Weitere
Lipide, wie die Hopanoide und Glykolipide, sind ebenfalls weit verbreitete
Bestandteile von bakteriellen Membranen (Hermans et al., 1991; Livermore &
Johnson, 1974).
Die Komponenten der Membran befinden sich in einem bestimmten Gleichgewicht
zueinander, um eine optimale Anpassung an den jeweiligen Lebensraum des
Organismus zu gewährleisten. Verändern sich die Umweltbedingungen für einen
Organismus, so kann dies zu Umstrukturierungen der Membranen führen. Unter
Phosphat-limitierenden Bedingungen werden die Phospholipide in
Sinorhizobium meliloti durch Phosphat-freie Lipide wie das Betainlipid DAG-N,N,N-
trimethylhomoserin ersetzt (Geiger et al., 1999). Ethanoltolerante Mikroorganismen
adaptieren durch einen verstärkten Einbau von langen, ungesättigten Fettsäuren an
Ethanolstress (Hermans et al., 1991).
Einleitung
5
Vorkommen von Phosphatidylcholin Genomsequenzanalysen bekannter Bakterienarten zeigten, dass etwa 10 % aller
bisher identifizierten Bakterien grundsätzlich zur PC-Biosynthese befähigt sind
(Sohlenkamp et al., 2003). Dabei lässt sich eine Verteilung in phylogenetisch weit
voneinander entfernten Bakteriengruppen ausmachen (Abb. 3). So besitzen u.a. die
Gram-negativen α-Proteobakterien Rhodobacter sphaeroides (Arondel et al., 1993),
Agrobacterium tumefaciens (Kaneshiro & Law, 1964), Bradyrhizobium japonicum
(Tang & Hollingsworth, 1998), S. meliloti (de Rudder et al., 1997), das γ-
Proteobakterium Pseudomonas aeruginosa (Wilderman et al., 2002) und das in die
Klasse der Spirochäten gehörende Bakterium Treponema denticola (Kent et al.,
2004) die genetische Information zur PC-Biosynthese (Abb. 3).
Abb. 3: Verbreitung von Phosphatidylcholin-Biosyn-thesegenen in Bakterien (modifiziert nach Sohlenkamp, et al., 2003). Phylogenetischer Stammbaum der Bakterien basierend auf vergleichenden 16SrRNA Sequenzanalysen. Gezeigt ist das Vorhandensein (+) oder das Fehlen (-) von Phosphatidylcholin-Biosynthese-genen in bakteriellen Genomen. Die typischen Modellorganismen der Mikrobiologie B. subtilis und E. coli sowie das in dieser Arbeit im Mittelpunkt stehende α-Proteobakterium A. tumefaciens sind in roter Schrift hervor gehoben.
In den letzten Jahren wurde für zehn Organismen PC als Membrankomponente
experimentell nachgewiesen. Dabei reicht der prozentuale PC-Anteil von 5,3 % der
Gesamtlipide in Zymomonas mobilis bis 74,3 % in Acetobacter aceti (Tab. 1),
wodurch die hohe Varianz in der Zusammensetzung bakterieller Membranen
verdeutlicht wird.
Einleitung
6
Tab. 1: Experimentell nachgewiesener PC-Anteil in verschiedenen Bakterien.
Organismus PC-Level Referenz
Acetobacter aceti 74,3 % (Hanada et al., 2001)
Agrobacterium tumefaciens ~ 23 % (Wessel et al., 2006);
(Hacker unveröffentlicht)
Brucella abortus 26,8 % (Comerci et al., 2006)
Bradyrhizobium japonicum 52,0 % (Minder et al., 2001)
Rhodobacter sphaeroides 27,1 % (Arondel et al., 1993)
Sinorhizobium meliloti 36,5 % (Sohlenkamp et al., 2000)
Zymomonas mobilis 5,3 % (Tahara et al., 1994)
Legionella pneumophila ~ 32 % (Conover et al., 2008)
Pseudomonas aeruginosa n.q. (Wilderman et al., 2002)
Treponema denticola ~ 60 % (Kent et al., 2004)
n.q., nicht quantifiziert
Eine mögliche Verteilung von PC auf die innere und äußere Membran Gram-
negativer Bakterien ist oft noch ungeklärt. Lediglich für Legionella pneumophila
(Hindahl & Iglewski, 1984) und P. aeruginosa (Wilderman et al., 2002) wurde gezeigt,
dass PC sowohl in der inneren als auch in der äußeren Membran lokalisiert ist. Ob
eine asymmetrische Verteilung von PC innerhalb einer Membran, vergleichbar zu der
eukaryotischen Plasmamembran, vorkommt, ist nicht bekannt.
Funktionen von Phosphatidylcholin Über die Funktionen von PC in Prokaryoten ist im Gegensatz zu der Rolle von PC in
Eukaryoten wenig bekannt. PC scheint jedoch nicht von essentieller Bedeutung für
prokaryotische Organismen zu sein, da PC-defiziente Mutanten in A. aceti,
A. tumefaciens, Brucella abortus, L. pneumophila, P. aeruginosa, R. sphaeroides,
S. meliloti und Z. mobilis lebensfähig sind (Arondel et al., 1993; Comerci et al., 2006;
Conde-Alvarez et al., 2006; Conover et al., 2008; de Rudder et al., 2000; Hanada et
al., 2001; Tahara et al., 1994; Wessel et al., 2006; Wilderman et al., 2002). Dabei
sind für viele der bisher untersuchten PC-Biosynthesemutanten Phänotypen
beschrieben, die auf wichtige und vielfältige Funktionen von PC in der Zelle
hindeuten. Beispielsweise weist eine PC-defiziente S. meliloti-Mutante selbst in
Anwesenheit des für die PC-Biosynthese wichtigen Cholins einen drastischen
Einleitung
7
Wachstumsdefekt auf (de Rudder et al., 2000). Die Abwesenheit von PC in
L. pneumophila führt zu einem Motilitätsverlust, der auf einer fehlenden Akkumulation
der Flagellinproteine beruht (Conover et al., 2008). In A. aceti wird ein
Zusammenhang zwischen der PC-Biosynthese und der Toleranz gegenüber Azidität
postuliert (Hanada et al., 2001).
Viele der in Abbildung 3 und Tabelle 1 aufgeführten PC-synthetisierenden Bakterien
gehen symbiontische oder pathogene Interaktionen mit einem eukaryotischen Wirt
ein. Infolgedessen belegen die Studien der letzten Jahre eine besondere Bedeutung
von PC für die Bakterien-Wirt-Interaktion. Dies wurde erstmals für B. japonicum, den
Stickstoff-fixierenden Wurzelsymbionten der Sojabohne Glycine max, gezeigt. Eine
PC-Biosynthesemutante von B. japonicum weist drastische symbiontische Defekte
auf. So zeigt das Innere der Wurzelknöllchen keine rötliche Färbung, was auf einen
Leghämoglobinmangel hindeutet. Die Anzahl der Bakteroide in den Wurzelknöllchen
ist deutlich reduziert und die Sojabohnen tragen gelbliche Blätter, was ein Anzeichen
für Stickstoffmangel ist. Dementsprechend liegt die Stickstofffixierungsaktivität bei
nur 18 % im Vergleich zum Wildtyp (Minder et al., 2001). In einem
Mausinfektionsmodell mit B. abortus PC-Biosynthesemutanten wird ein drastischer
Virulenzdefekt durch die fehlende Etablierung einer Replikationsnische in
Makrophagen deutlich (Comerci et al., 2006; Conde-Alvarez et al., 2006). Außerdem
weist eine PC-defiziente Mutante des Humanpathogens L. pneumophila eine
geringere Cytotoxizität und Besiedlungsdichte gegenüber Makrophagen auf. Dieser
Virulenzdefekt ist mit einer verminderten Adhäsion, einem nicht funktionalen Typ IV-
Sekretionssystems (TIVSS) und dem bereits erwähnten Motilitätsverlust assoziiert
(Conover et al., 2008). Des Weiteren ist das Pflanzenpathogen A. tumefaciens in
Abwesenheit von PC nicht mehr in der Lage, eine Tumorbildung in der Pflanze zu
induzieren (Wessel et al., 2006). Die Ursachen dieses Phänotyps werden in Kapitel
3.3 näher diskutiert.
Die molekularen Gründe für die aufgeführten Phänotypen der PC-
Biosynthesemutanten sind in den meisten Fällen noch ungeklärt. Jedoch lassen
erste Hinweise darauf schließen, dass die Rolle, die PC in der Interaktion mit einem
eukaryotischen Wirt spielt, zwar von entscheidender, aber unterschiedlicher
Bedeutung ist.
Einleitung
8
2. Biosynthese von Phosphatidylcholin
Sowohl in Eu- als auch in Prokaryoten wird PC über verschiedene Biosynthesewege
gebildet, wobei es signifikante Unterschiede zwischen den beiden Domänen gibt.
Ebenso variiert die Gewichtung der Synthesewege von Organismus zu Organismus
bzw. von Gewebe zu Gewebe in einem Lebewesen.
2.1 Biosynthese von Phosphatidylcholin in Eukaryoten
Eukaryoten synthetisieren PC hauptsächlich über den Methylierungsweg und den
CDP-Cholinweg, auch „Kennedyweg“ genannt. Bei der PC-Biosynthese über den
Methylierungsweg erfolgt eine dreifache Methylierung des Vorläufermoleküls PE
de novo über die Intermediate MMPE und DMPE zu PC, wobei S-Adenosylmethionin
(SAM) als Methylgruppen-Donor dient (Abb. 4A). Die Methylierungsreaktionen
werden durch Phospholipid N-Methyltransferasen (Pmt) katalysiert (Bremer &
Greenberg, 1959; Vance & Ridgway, 1988). In Säugetieren katalysiert ein Pmt-
Enzym (PEMT) alle drei Methylierungsreaktionen. Dabei existieren zwei Isoformen
des Enzyms, die sich in ihrer intrazellulären Lokalisation unterscheiden. PEMT1
kommt im Endoplasmatischen Reticulum vor und ist für die Hauptaktivität
verantwortlich, während PEMT2 in Mitochondrien-assoziierten Membranen lokalisiert
ist (Walkey et al., 1997). Im Gegensatz zu den Säugetieren sind in den Hefen
Saccharomyces cerevisiae und Schizosaccharomyces pombe zwei verschiedene
Pmt-Varianten notwenig, um PE in PC zu überführen. Die in Klasse I und II
eingeteilten Pmt-Enzyme unterscheiden sich hinsichtlich ihrer Substratspezifität. Die
erste Methylierung von PE zu MMPE wird durch Klasse II-Pmt’s und die letzten
beiden Methylierungsreaktionen von MMPE zu PC (S. pombe) oder alle drei
Methylierungen (S. cerivisiae) durch Klasse I-Pmt’s katalysiert (Kanipes & Henry,
1997).
Die Herstellung von PC über den CDP-Cholinweg beruht auf einem Transfer der
Phosphocholingruppe von CDP-Cholin auf DAG (Abb. 4B). Die daran beteiligten
Enzyme umfassen eine Cholin-Kinase (CKI), welche freies Cholin phosphoryliert,
und eine CTP:Phosphocholin-Cytidylyltransferase (CCT), um das Cholinphosphat in
CDP-Cholin umzuwandeln. Die darauf folgende Verbindung von CDP-Cholin mit
Einleitung
9
DAG wird durch eine 1,2-Diacylglycerin-Cholin-Phosphotransferase (CPT) katalysiert
(Kennedy & Weiss, 1956; Weiss et al., 1958). Da PC einem ständigen Abbau durch
Phospholipasen unterliegt, können die aus der Hydrolyse entstehenden Produkte für
eine erneute PC-Synthese genutzt werden, so dass über den CDP-Cholinweg eine
ständige Wiederaufbereitung der PC-Moleküle möglich ist (Billah & Anthes, 1990;
Exton, 1994). Das für den CDP-Cholinweg essentielle Cholin kann entweder durch
die PC-Hydrolyse entstehen oder als exogenes Cholin in die Zelle transportiert
werden (Kent, 1990).
A) Methylierungsweg B) CDP-Cholinweg
Abb. 4: Phosphatidylcholin-Biosynthese in Eukaryoten (modifiziert nach Sohlenkamp, et al., 2003). A) Die drei Methylierungen an der Stickstoffgruppe (roter Kreis) werden durch Phospholipid N-Methyltransferasen (Pmt) katalysiert, wobei S-Adenosylmethionin (SAM) als Methylgruppen-Donor dient und zu S-Adenosylhomocystein (SAH) umgesetzt wird. B) Über den CDP-Cholinweg wird PC, basierend auf dem Transfer des Phosphocholins von CDP-Cholin auf DAG, synthetisiert. Dabei unterliegt PC einem ständigen Abbau durch Phospholipasen (PLC/PLD). ADP, Adenosindiphosphat; ATP, Adenosintriphosphat; CCT, CTP:Phosphocholin-Cytidylyltransferase; CDP, Cytidindiphosphat; CKI, Cholin-Kinase; CMP, Cytidinmonophosphat; CPT, 1,2-Diacylglycerin-Cholin-Phosphotransferase; CTP, Cytidintriphosphat; DAG, Diacylglycerin; Pi, Phosphatgruppe.
Einleitung
10
Neben den beiden Hauptsynthesewegen, können PC-Moleküle in Eukaryoten über
zwei weitere Wege entstehen. Zum einen kann PC durch den Austausch von
Kopfgruppen gebildet werden (Kanfer, 1980) und zum anderen über die Acetylierung
von Lyso-PC (Hatch et al., 1989).
Innerhalb der Domäne der Eukaryoten unterscheidet sich die Gewichtung der PC-
Biosynthesewege deutlich. Pilze und Hefen produzieren PC hauptsächlich über den
Methylierungsweg (Kanipes & Henry, 1997), während in Säugetieren und Pflanzen
vorwiegend der CDP-Cholinweg genutzt wird (Dewey et al., 1994; Walkey et al.,
1997).
Regulation der Phosphatidylcholin-Biosynthese Da PC von großer Bedeutung ist und vielfältige Funktionen in Eukaryoten erfüllt, gibt
es umfassende Regulationsmechanismen, welche die PC-Biosynthese kontrollieren.
In S. cerevisiae wird die Pmt-Aktivität des Methylierungsweges hauptsächlich auf
transkriptioneller Ebene reguliert. Die beteiligten Gene werden durch die
Verfügbarkeit von Inositol und Cholin, sowie Wachstumsphasen-abhängig
kontrolliert. Das Vorkommen von Inositol und Cholin führt zu einer starken
Repression beider Pmt-kodierenden Gene. Die Zugabe von Cholin allein hat jedoch
keinen Einfluss auf die Expression (Gaynor et al., 1991; Kodaki et al., 1991). Durch
den Eintritt in die stationäre Wachstumsphase werden die Pmt-Aktivitäten um den
Faktor 2-5 reprimiert (Lamping et al., 1994). In Säugetieren ist die PEMT-Aktivität
von den Konzentrationen der Substrate PE und SAM, sowie dem Produkt SAH
abhängig. Auf transkriptioneller Ebene wird die Expression von pemt während der
Entwicklung und durch die Cholin-Versorgung kontrolliert. Dabei erfolgt die
Regulation der Methylierungs- und CDP-Cholinwege reziprok (Ridgway & Vance,
1988; Ridgway et al., 1989).
Der CDP-Cholinweg wird hauptsächlich durch die CCT, welche Cholinphosphat in
CDP-Cholin umsetzt, als geschwindigkeitsbestimmende Komponente beeinflusst.
Die Aktivität der CCT wird durch verschiedene Phospholipide induziert und ist von
dem Transfer des Enzyms aus dem Cytosol zur Membran abhängig (Kent, 1990).
Auf transkriptioneller Ebene wird die Expression der CCT durch den Zellzyklus, das
Zellwachstum sowie der Zelldifferenzierung beeinflusst. Dabei induzieren die
Transkriptionsfaktoren Sp1, Sp3, Rb, TEF4, Ets-1 und E2F die Expression, während
der Faktor Net diese reprimiert (Li & Vance, 2008). Bisher geht man davon aus, dass
Einleitung
11
die beiden anderen Enzyme des CDP-Cholinwegs, CKI und CPT, in der Regulation
nur eine untergeordnete Rolle spielen.
2.2 Biosynthese von Phosphatidylcholin in Prokaryoten
Die erste bakterielle Pmt-Aktivität wurde 1964 von Kaneshiro und Law an zellfreien
Extrakten aus A. tumefaciens demonstriert (Kaneshiro & Law, 1964). Erst viele Jahre
später wurden in R. sphaeroides und Z. mobilis die für Pmt-Enzyme kodierenden
Gene identifiziert, isoliert und charakterisiert (Arondel et al., 1993; Tahara et al.,
1994). Allerdings konnte keine enzymatische Aktivität des CDP-Cholinwegs
detektiert werden, so dass nur ein existierender Methylierungsweg für die PC-
Biosynthese in Bakterien postuliert wurde. Im Jahr 1999 wurde eine pmt-Mutante in
S. meliloti untersucht, die keine Intermediate MMPE und DMPE aufweist, jedoch PC
in Wildtyp-Mengen in Anwesenheit von Cholin im Medium synthetisierte.
Infolgedessen wurde ein neuer PC-Biosyntheseweg (PC-Synthaseweg) in S. meliloti
entdeckt (de Rudder et al., 1999), so dass auch in Bakterien grundsätzlich zwei
verschieden Wege zur PC-Produktion existieren.
Der Methylierungsweg Vergleichbar zu den Eukaryoten wird PC in Prokaryoten durch eine dreifache
Methylierung von PE über MMPE und DMPE mit SAM als Methylgruppen-Donor
synthetisiert (Abb. 5). Die für die Methylierungsreaktionen erforderlichen Pmt-
Enzyme unterscheiden sich deutlich in Sequenz und Struktur von den
eukaryotischen Enzymen, so dass kaum homologe Bereiche vorliegen (Sohlenkamp
et al., 2003).
Das aus R. sphaeroides isolierte Pmt-Enzym produziert nach heterologer Expression
in E. coli PC, jedoch nicht die Intermediate MMPE und DMPE (Arondel et al., 1993).
Die heterologe Expression des Pmt-Enzyms aus S. meliloti in E. coli führt neben der
PC-Synthese zu signifikanten MMPE- und geringen DMPE-Mengen (de Rudder et
al., 2000). Ein Sequenzvergleich beider Enzyme weist nur eine geringe Ähnlichkeit
auf, die sich auf einen kurzen Sequenzabschnitt um das Motiv VLE/DXGXGXG
bezieht. Dieses Motiv ist charakteristisch für Methyltransferasen und an der Bindung
des Methylgruppen-Donors SAM beteiligt (Haydock et al., 1991). Aufgrund der
Einleitung
12
Unterschiede werden die Pmt-Enzyme aus R. sphaeroides und S. meliloti in zwei
verschiedene Familien eingeordnet. Der Rhodobacter Pmt-Typ mit weiteren
Vertretern aus L. pneumophila und A. aceti zeigt Ähnlichkeiten zu
Ubiquinon/Menaquinon-Biosynthese-Methyltransferasen (UbiE) (Hanada et al., 2001;
Martínez-Morales et al., 2003). Dagegen weist der Sinorhizobium Pmt-Typ, der auch
Pmt-Proteine aus A. tumefaciens und B. abortus umfasst (Comerci et al., 2006;
Conde-Alvarez et al., 2006; Wessel et al., 2006), Homologien zu rRNA-Methylasen
auf (Sohlenkamp et al., 2003).
Methylierungsweg PC-Synthaseweg
Abb. 5: Phosphatidylcholin-Biosynthese in Prokaryoten. Schematische Darstellung einer Bakterienzelle mit den im Text beschriebenen PC-Biosynthesewegen (Methylierungs- und PC-Synthaseweg). Die drei Methylierungsreaktionen finden an dem Stickstoff der polaren Kopfgruppe (roter Kreis) der Lipide statt und werden durch Phospholipid N-Methyltransferasen (Pmt) katalysiert. Die direkte Kondensation von Cholin mit CDP-Diacylglycerin zu PC wird durch die membranständige PC-Synthase (Pcs) katalysiert. Es wird postuliert, dass Cholin oder Vorläufermoleküle von einigen Bakterien durch zum größten Teil noch nicht identifizierte Transporter aufgenommen werden. CDP, Cytidindiphosphat; CMP, Cytidinmonophosphat; SAH, S-Adenosylhomocystein; SAM, S-Adenosylmethionin.
Im Gegensatz zu den bisher aufgeführten Organismen, die ein einziges pmt-Gen
besitzen, wurden in B. japonicum fünf pmt-Gene identifiziert (pmtA und pmtX1-X4).
Einleitung
13
Die Enzyme PmtA, PmtX3 und PmtX4 werden dem Sinorhizobium Pmt-Typ
zugeordnet und PmtX1 sowie PmtX2 gehören in die Familie der Rhodobacter Pmt’s.
Aktivitätsanalysen der Pmt-Proteine in E. coli belegen unterschiedliche
Substratspezifitäten für diese Enzyme. In B. japonicum Wildtyp lassen sich
Aktivitäten für PmtA und PmtX1 nachweisen. Dabei katalysiert PmtA die erste
Methylierung von PE zu MMPE und PmtX1 die nachfolgenden Methylierungen über
DMPE zu PC (Hacker et al., 2008b; Minder et al., 2001).
Über die Regulation und Aktivität der bakteriellen Pmt-Emzyme ist im Vergleich zu
den eukaryotischen Proteinen wenig bekannt. Die biochemische Charakterisierung
einer Pmt aus A. tumefaciens (PmtA) gibt erste Hinweise über die Regulation der
Aktivität und den Reaktionsmechanismus. Diese kann in vitro alle drei Substrate (PE,
MMPE und DMPE) zur PC-Synthese nutzen, wobei die Verfügbarkeit eines der
Lipidsubstrate für die Bindung des Methylgruppen-Donors SAM essentiell ist. Die
Aktivität wird durch die Produkte PC und S-Adenosylhomocystein (SAH) inhibiert
(Aktas & Narberhaus, 2009).
Der Phosphatidylcholin-Synthaseweg Im Unterschied zu dem CDP-Cholinweg der Eukaryoten wird über den bakteriellen
PC-Synthaseweg Cholin direkt mit CDP-Diacylglycerin zu PC kondensiert (Abb. 5).
Als sekundäres Produkt entsteht dabei Cytidinmonophosphat (CMP). Diese Reaktion
wird durch die membranständige PC-Synthase (Pcs) katalysiert, die ausschließlich in
Bakterien vorkommt (de Rudder et al., 1999). Für Pcs-Enzyme werden 6-8
Transmembranhelices vorhergesagt und signifikante Sequenzähnlichkeiten zu
anderen CDP-Alkohol-Phosphatiydltransferasen sowie der Phosphatidylserin-
synthase (Pss) (Sohlenkamp et al., 2000; Sohlenkamp et al., 2003).
Es wird diskutiert, dass Cholin oder Vorläufermoleküle von einigen Bakterien über
Transporter aufgenommen werden, um Cholin für den PC-Synthaseweg zur
Verfügung zu stellen. Die mit einem eukaryotischen Wirt interagierenden Bakterien
scheinen dabei durch den Wirt mit Cholin versorgt zu werden (Martínez-Morales et
al., 2003). Beispielsweise stellt die Wirtspflanze Cholin für den Leguminosen-
Symbionten S. meliloti bereit (de Rudder et al., 1999). Dabei wird Cholin u.a. über
den hochaffinen ABC- (ATP-binding cassette) Transporter (Cho) in die Bakterienzelle
aufgenommen (Dupont et al., 2004). Des Weiteren ist die PC-Synthese über den PC-
Synthaseweg in B. abortus vom Cholin des Wirtes abhängig (Comerci et al., 2006).
Einleitung
14
Die tierpathogenen P. aeruginosa, Borrelia burgdorferi, L. pneumophila und
Brucella melitensis sowie die pflanzenpathogenen A. tumefaciens und
Pseudomonas syringae besitzen pcs-Gene und somit die Möglichkeit, den Cholin-
abhängigen PC-Synthaseweg zu nutzen (Sohlenkamp et al., 2003). Dabei könnte es
für diese Organismen von Vorteil sein, PC über den PC-Synthaseweg mit zur
Verfügung gestelltem Cholin zu bilden, um den energieaufwendigeren
Methylierungsweg (3 SAM-Moleküle pro Molekül PC) zu umgehen.
Die Bereitstellung von Cholin für den PC-Synthaseweg könnte auch durch die
membranständigen Pcs-Enzyme vermittelt werden. Allerdings gibt es noch keine
experimentellen Belege für eine direkte Verwendung des Cholins aus dem
Periplasma durch Pcs.
Erste Hinweise auf biochemische Eigenschaften von Pcs-Enzymen liefern in vitro
Experimente mit S. meliloti Rohextrakten (de Rudder et al., 1999). S. meliloti Pcs
besitzt ein pH Optimum von 8,0 und ist abhängig von bivalenten Kationen sowie dem
zwitterionischen Detergenz Triton X-100. Dabei wurde eine optimale Pcs-Aktivität in
Gegenwart von 10 mM MnCl2 und 0,2 % (w/v) Triton X-100 detektiert. Bei einer
Cholinkonzentration von < 50 mM erreicht die Pcs-Aktivität typische
Sättingungskinetiken, wenn das molare Verhältnis von CDP-Diacylglycerin variiert.
Die Kurve verläuft sigmoidal bei einer Cholinkonzentration von über 100 mM,
wodurch entweder ein positiver Effekt durch CDP-Diacylglycerin postuliert werden
kann oder eine Inhibierung durch das Substrat Cholin (de Rudder et al., 1999).
Nicht alle Prokaryoten besitzen beide PC-Biosynthesewege und auch die
Gewichtung der Synthesewege ist von Organismus zu Organismus verschieden. In
A. aceti und R. sphaeroides wird PC nur über den Methylierungsweg synthetisiert
(Arondel et al., 1993; Hanada et al., 2001), während in B. abortus, B. burgdorferi und
P. aeruginosa nur der PC-Synthaseweg funktionell ist (Comerci et al., 2006;
Martínez-Morales et al., 2003; Wilderman et al., 2002). In A. tumefaciens, S.meliloti
und B. japonicum konnten beide PC-Biosynthesewege detektiert werden (de Rudder
et al., 1999; de Rudder et al., 1997; Hacker et al., 2008b; Minder et al., 2001; Wessel
et al., 2006).
Einleitung
15
3. Der Modellorganismus Agrobacterium tumefaciens
Das Gram-negative α-Proteobakterium A. tumefaciens ist ein stäbchenförmiges,
bewegliches Mitglied der Familie Rhizobiacea. Das 2001 vollständig sequenzierte
Genom mit einer Größe von 5,67 Mbp besteht aus einem linearen und einem
zirkulären Chromosom sowie zwei Plasmiden (pAtC58 und pTiC58/Ti-Plasmid)
(Goodner et al., 2001; Wood et al., 2001). Im Gegensatz zu anderen Vertretern
dieser Familie ist A. tumefaciens nicht zur Stickstofffixierung befähigt. Es handelt sich
um einen pflanzenpathogenen Organismus, der mit einem breiten Wirtsspektrum
dikotyle Pflanzen infiziert und eine Tumorbildung im Pflanzengewebe induziert.
3.1 Der Infektionsprozess
Für eine erfolgreiche Infektion durch A. tumefaciens muss ein Teil der bakteriellen
DNA, die sog. T-DNA, in die Pflanzenzelle transferiert werden, um dort in den
Zellkern transportiert und letztendlich in das Genom integriert zu werden. Die T-DNA
ist ein Abschnitt des Ti-Plasmids („tumor-inducing plasmid“), der von spezifischen,
25 bp langen Sequenzen flankiert ist (Stachel & Nester, 1986). Neben der T-DNA
befinden sich die in acht Abschnitten organisierten Virulenzgene (virA-H) auf dem Ti-
Plasmid, deren Genprodukte u.a. für die Prozessierung und den Transfer der T-DNA
benötigt werden. Die Expression der vir-Gene wird durch das Zwei-
Komponentensystem VirA/G reguliert (Abb. 6) (Jin et al., 1990; Stachel & Zambryski,
1986). Die Sensorkinase VirA ist in der inneren Membran lokalisiert und wird durch
eine Autophosphorylierung an einem konservierten Histidinrest aktiviert. Dabei
induzieren die im Wundsekret verletzter Pflanzen enthaltenen phenolischen
Verbindungen und Monosaccharide als Signalsubstanzen sowie ein externer saurer
pH die Phosphorylierung (McLean et al., 1994). VirA gibt den Phosphatrest an einen
konservierten Aspartatrest des cytoplasmatischen Antwortregulators VirG weiter, so
dass dieser aktiviert wird (Jin et al., 1990). Als aktives Protein bindet VirG an die
12 bp langen vir-Boxen der vir-Promotoren, um die Expression der vir-Gene zu
induzieren (Stachel et al., 1986; Winans, 1992). Die virB1-11/D4-Gene kodieren für
ein TIVSS, welches die innere und äußere Membran durchspannt und einen langen
Pilus für den Durchtritt von Makromolekülen bildet (Beijersbergen et al., 1994; Zupan
Einleitung
16
& Zambryski, 1995). Aufgrund der Funktion und Lokalisation lassen sich die Proteine
des TIVSS’s in drei Subkomplexe unterteilen. VirB4, VirB11 und VirD4 besitzen ein
Walker A-Motiv und sind somit durch ATP-Bindung und -Hydrolyse für die
Energiezufuhr verantwortlich. Die Proteine VirB6-10 bilden den Kernkomplex und
VirB2-3 sowie VirB5 den Pilus (Baron et al., 2002; Christie, 1997; Krall et al., 2002b).
Abb. 6: Virulenzkaskade in A. tumefaciens. Die Expression der vir-Gene wird durch das Zwei-Komponentensystem VirA/G reguliert. Die Autophosphorylierung der Sensorkinase VirA wird durch phenolische Verbindungen (z.B. Acetosyringon), Azidität [H+] und Monosaccharide (mittels ChvE) induziert. VirA gibt den Phosphatrest (P) an den Antwortregulator VirG weiter, der in seiner aktiven Form an die vir-Boxen der vir-Genpromotoren bindet und so die Transkription induziert. Die prozessierte T-DNA und VirE2 Proteine werden über das Typ IV Sekretionssystem (TIVSS), bestehend aus den VirB und VirD4 Proteinen, in die Pflanzenzelle transferiert. ADP, Adenosindiphosphat; ATP, Adenosintriphosphat; ÄM, äußere Membran; D, Aspartat; H, Histidin; IM, innere Membran.
Durch die Kombination von biochemischen und genetischen Analysen wurde ein
Modell zur Assemblierung des TIVSS’s entwickelt. Dabei initiiert VirB8 die
Assemblierung und führt VirB1 zum Zellpol, wo es wahrscheinlich die Zellwand lokal
lysiert, um dort den Aufbau des Kernkomplexes zu ermöglichen. Unmittelbar nach
der Bildung des Kernkomplexes, werden die Proteine VirB2, VirB3 und VirB5 für den
Pilusaufbau rekrutiert. Dabei liefert die ATPase VirB11 die Energie für die VirB2-
Einleitung
17
Polymerisation durch das Periplasma (Christie & Cascales, 2005; Judd et al., 2005;
Yuan et al., 2005).
Nach erfolgreicher Assemblierung des TIVSS’s binden VirD1 (Topoisomerase) und
VirD2 (Relaxase) als Komplex an die flankierenden Sequenzen der T-DNA, um diese
durch einen Einzelstrangbruch freizusetzen. Am 5’-Ende der T-DNA bleibt VirD2
kovalent gebunden und wird, ebenso wie das Einzelstrang-bindende Protein VirE2,
über das TIVSS in die Pflanzenzelle transferiert. VirE2 bindet an die einzelsträngige
T-DNA, um diese in der Pflanzenzelle vor einem nukleolytischen Abbau zu schützen
(Zupan & Zambryski, 1995). Außerdem vermittelt VirE2 den Transport der T-DNA in
den Nukleus über das nukleäre Importsystem durch die Bindung des zellulären VIP1
Proteins und VirD2 durch Interaktion mit Karyopherin α (Djamei et al., 2007). Nach
Integration der T-DNA in das Genom der Pflanzenzelle werden die bakteriellen Gene
(Onkogene) der T-DNA transkribiert. Die Onkogene kodieren für Enzyme, die in die
Biosynthese der Phytohormone Auxin und Cytokinin involviert sind, so dass es zu
unkontrollierten Zellteilungen und letztendlich zur Tumorbildung kommt (Krall et al.,
2002a; Ziemienowicz, 2001). Außerdem werden Gene auf der T-DNA kodiert, deren
Genprodukte für die Synthese sog. Opine verantwortlich sind. Diese werden durch
das Tumorgewebe sekretiert und von den A. tumefaciens-Zellen als Kohlenstoff- und
Stickstoffquelle genutzt. Die Gene, die die Aufnahme- und Abbausysteme der Opine
kodieren, sind auf dem in der Zelle verbleibenden Ti-Plasmid lokalisiert. Die Pflanzen
hingegen besitzen die entsprechenden Gene nicht und können somit die Opine nicht
nutzen. Je nach Art des synthetisierten Opins lassen sich die Ti-Plasmide in
verschiedene Klassen unterteilen. Beispiele hierfür sind die Nopalin und Octopin
Typ-Plasmide. Nopalin wird durch Kondensation von Arginin und α-Ketoglutarat
hergestellt, wobei Octopin eine Verbindung aus Arginin und Pyruvat darstellt
(Montoya et al., 1977; Ziemienowicz, 2001).
Mittels rekombinanter DNA-Techniken ist es möglich, die Onkogene durch jede
beliebige, fremde DNA zu ersetzen. Wenn die flankierenden Sequenzen der T-DNA
erhalten bleiben, wird die Fremd-DNA in die Pflanzenzelle transferiert und dort in das
Genom integriert (Gelvin, 2003). In der Pflanzenbiotechnologie wird eine Vielzahl
modifizierter Ti-Plasmide für die Erzeugung transgener Pflanzen eingesetzt, so dass
A. tumefaciens für die Landwirtschaft von zentraler Bedeutung ist.
Einleitung
18
3.2 Virulenzfaktoren
Neben den auf dem Ti-Plasmid lokalisierten vir-Genen wurden in A. tumefaciens
mittlerweile auch weitere, chromosomal kodierte Virulenzfaktoren identifiziert. Einige
der Faktoren beeinflussen die Regulation des Zwei-Komponentensystems VirA/G,
die Adhäsion an die Wirtsoberfläche oder die Expression der virC/D-Gene. Für viele
dieser Faktoren ist der molekulare und/oder funktionelle Zusammenhang noch nicht
geklärt.
Feinregulation des Zwei-Komponentensystems VirA/G Für eine optimale Induktion der vir-Genexpression muss das Zwei-
Komponentensystem VirA/G verschiedene Stimuli wahrnehmen, verarbeiten und
koordinieren. Phenolische Verbindungen sind die Hauptstimuli der Virulenzkaskade
und werden an der cytoplasmatischen Linkerdomäne von VirA erkannt (Brencic et
al., 2004; Melchers et al., 1989). Monosaccharide und Azidität sind als weitere
Signale für die Feinregulation verantwortlich und werden durch chromosomal
kodierte Faktoren vermittelt.
Das periplasmatische Protein ChvE weist signifikante Homologien zu vielen
bekannten Zucker-bindenden Proteinen auf. ChvE bindet Monosaccharide und
vermittelt die Zucker-abhängige Aktivierung der Sensorkinase VirA über die
periplasmatische Domäne (Abb. 6) (Cangelosi et al., 1990).
Das Zwei-Komponentensystem ChvG/I ist in der VirA-unabhängigen Regulation von
VirG involviert. Die Expression von VirG wird durch zwei Promotoren kontrolliert. Der
Promotor P1 wird durch phenolische Verbindungen und Phosphatmangel reguliert
und der Promotor P2 durch das Zwei-Komponentensystem ChvG/I. Man geht von
folgendem Modell aus: der Promotor P2 wird bei einem sauren pH mittels ChvG/I
aktiviert, so dass eine adäquate Menge an VirG vorhanden ist. Die durch
phenolische Verbindungen aktivierte Sensorkinase VirA phosphoryliert VirG, welches
autoregulativ die Transkription des eigenen Gens und der anderen vir-Gene induziert
(Charles & Nester, 1993; Li et al., 2002; Mantis & Winans, 1993; Winans, 1990).
Adhäsion von A. tumefaciens an Wirtsoberflächen Durch eine Tn5-Mutagenese wurden bereits 1980 die chromosomal kodierten Gene
chvA und chvB als wichtige Komponenten der Virulenz identifiziert (Garfinkel &
Einleitung
19
Nester, 1980). Der Virulenzdefekt der chvA- und chvB-Mutanten beruht auf einer
fehlerhaften Synthese, Akkumulation und Sekretion von β-(1,2)-Glukanen, die für die
Adhäsion der Bakterien an die Pflanzenoberfläche wichtig sind (de Iannino & Ugalde,
1989; Douglas et al., 1985).
Ein weiterer, wichtiger Faktor für die Adhäsion und eine nachfolgende
dreidimensionale Kolonisierung auf Oberflächen (Biofilmbildung) ist die Motilität der
Zellen. A. tumefaciens-Mutanten (ΔflgE, Δmot, ΔflaABC), die keine oder nur
fehlerhafte Flagellen ausbilden und somit einen drastischen Motilitätsdefekt
aufweisen, sind in der Adhäsion und Biofilmbildung stark beeinträchtigt. Als Folge
dessen zeigen diese Mutanten auch eine verminderte Virulenz (Chesnokova et al.,
1997; Merritt et al., 2007). Die Transkription des flaABC-Operons und somit die
Anzahl der Flagellen lässt sich durch das Wachstum von A. tumefaciens-Zellen im
Dunkeln erhöhen. Sowohl die Adhäsion an Wurzeln der Tomatenpflanze als auch die
Tumorbildung am Spross sind bei Inkubation der Bakterien im Dunkeln erhöht. Damit
besteht ein Zusammenhang zwischen Licht und Virulenz, der durch den Einfluss auf
die Licht-abhängige Expression der Flagellenproteine und der Motilität vermittelt wird
(Oberpichler et al., 2008).
Duale Kontrolle der virC-, virD- und ipt-Gene Die virC/D-Genprodukte sind in die Prozessierung der T-DNA involviert und werden
durch das Zwei-Komponentensystem VirA/G positiv reguliert. Das Onkogen ipt ist auf
der T-DNA lokalisiert und wird dementsprechend in der Pflanzenzelle von der
Transkriptionsmaschinerie des Wirtes erkannt und gebildet. Das ipt-Gen kodiert für
ein Enzym der Cytokininbiosynthese und ist somit an der Tumorbildung beteiligt
(Kado, 2002). Die Expression der virC/D- und ipt-Gene im Bakterium wird durch den
Transkriptionsrepressor Ros negativ reguliert. Mutationen in ros führen zu einer
erhöhten Anzahl an T-DNA Zwischenprodukten in der Zelle, hervorgerufen durch die
Aufhebung der Repression von virC und virD (Close et al., 1987). Außerdem wird in
ros-Mutanten ipt konstitutiv exprimiert. Dies resultiert in einer Cytokininproduktion
innerhalb des Bakteriums (Chou et al., 1998).
Die Protease Lon und das kleine Hitzeschockprotein HspL Die ATP-anhängige Protease Lon ist ein konserviertes Protein, welches eine zentrale
Rolle in der Degradation fehlgefalteter Proteine und spezifischen Regulatoren in
Einleitung
20
vielen verschiedenen Organismen spielt (Gottesman & Zipser, 1978). Obwohl in
A. tumefaciens bisher noch keine Lon-Substrate identifiziert wurden, konnte durch
phänotypische Charakterisierung und Komplementationsexperimente einer lon-
Mutante eine Funktion der Protease in zellulären Prozessen zugeordnet werden. Lon
ist für die Zellteilung wichtig und ein Mitglied der Sigma32-abhängigen
Hitzeschockantwort. Darüber hinaus zeigt die lon-Mutante einen drastischen
Virulenzdefekt auf Kalanchoe diagremontiana-Blättern (Su et al., 2006). In den
anderen α-Proteobakterien B. abortus und S. meliloti wird Lon ebenfalls als eine
wichtige Komponente in der Bakterien-Wirts-Interaktion diskutiert (Robertson et al.,
2000; Summers et al., 2000). Allerdings sind die molekularen Zusammenhänge noch
nicht aufgeklärt.
Ein weiteres Mitglied der Hitzeschockantwort ist das kleine Hitzeschockprotein (Hsp)
HspL, welches in die Klasse des α-Kristallin-Typs eingeordnet wird. Neben HspL
wurden drei weitere Hsp-Proteine in A. tumefaciens identifiziert: HspC, HspAT1 und
HspAT2. Die Expression von HspL, HspAT1 und HspAT2 wird durch Hitzeschock
induziert, wobei die Regulation von HspL Sigma32-abhängig ist (Balsiger et al.,
2004). Interessanterweise konnte durch vergleichende Proteomstudien auch eine
Virulenz-abhängige Induktion der HspL-Expression beobachtet werden. Diese
deutlich induzierte HspL-Expression unter Virulenzbedingungen zeigt sich jedoch nur
im Wildtyp, nicht aber in virA- und virG-Mutanten (Abb. 7), so dass eine Regulation
über das Zwei-Komponentensystem VirA/G möglich wäre (Lai et al., 2006). Da in der
Promotorregion des hspL-Gens keine vir-Box Sequenzen für die Bindung des
Antwortregulators VirG vorliegen, ist die VirA/G-abhängige Expressionsinduktion
ebenso wie die Funktion von HspL in der Virulenz unklar.
Abb. 7: VirA/G-abhängige HspL-Expression (modifiziert nach Lai et al., 2006). Ausschnitt aus der Proteomanalyse mit Gesamtprotein-extrakt von A. tumefaciens Wildtyp und virA/G-Mutanten, die unter Virulenz-induzierten Bedingungen kultiviert wurden. VirK ist als Positivkontrolle für das vir-Regulon markiert.
Einleitung
21
3.3 Biosynthese und Bedeutung von Phosphatidylcholin in A. tumefaciens
A. tumefaciens synthetisiert PC über den Methylierungs- und den PC-Synthaseweg.
Basierend auf Genomsequenzanalysen wird ein Pmt-Enzym für den
Methylierungsweg und ein Pcs-Enzym für die Kondensation von Cholin mit CDP-
Diacylglycerin zu PC postuliert. Die beiden entsprechenden PC-Biosynthesegene,
pmtA und pcs, sind auf dem zirkulären Chromosom lokalisiert. Die Charakterisierung
von ∆pmtA- und ∆pcs-Mutanten sowie einer Doppelmutante (ΔpmtA Δpcs) bezüglich
ihrer Phospholipidzusammensetzung zeigte, dass PmtA und Pcs die einzigen PC-
Biosyntheseenzyme in A. tumefaciens sind (Wessel et al., 2006). Der Wildtyp bildet
die hauptsächlich vorkommenden Lipide PE, MMPE, DMPE und PC (Abb. 8). Da in
der ΔpmtA-Mutante nur der PC-Synthaseweg funktionell ist, werden PE und PC,
nicht aber die Zwischenstufen MMPE und DMPE, detektiert. Die Δpcs-Mutante zeigt
ein Wildtyp-ähnliches Lipidmuster, da PC über den Methylierungsweg synthetisiert
wird. Die Doppelmutante (ΔpmtA Δpcs) weist nur noch PE auf, d.h. dieser Stamm ist
nicht mehr in der Lage, PC zu produzieren (Wessel et al., 2006).
Abb. 8: Phospholipidspektrum von A. tumefaciens Wildtyp und PC-Biosynthesemutanten (modifiziert nach Wessel et al., 2006). Die Lipide der verschiedenen Stämme wurden nach Anzucht in YEB-Vollmedium bei 30°C extrahiert und mittels eindimensionaler Dünnschichtchromatographie aufgetrennt. Die Färbung der Phospholipide erfolgte mit „Molybdenum Blue Spray“ (Sigma-Aldrich). Schwach zu detektierende Lipide sind mit einem Pfeil markiert. DMPE, Dimethylphosphatidylethanolamin; MMPE, Monomethylphosphatidylethanolamin; PC, Phosphat-idylcholin; PE, Phosphatidylethanolamin.
Werden die Stämme in Minimalmedium ohne Cholinzugabe kultiviert, weist die
ΔpmtA-Mutante nur minimale PC-Mengen auf (Wessel et al., 2006). In Gegenwart
von Cholin ist die ΔpmtA-Mutante jedoch in der Lage, signifikante PC-Mengen über
den PC-Synthaseweg zu produzieren (Daten nicht gezeigt, Sonja Klüsener). Dies
deutet darauf hin, dass A. tumefaciens Cholin aus der Umgebung aufnimmt. Ein
putativer Cholin ABC-Transporter wird derzeit in unserer Arbeitsgruppe untersucht.
Die PC-Biosynthesemutanten wachsen sowohl in Voll- als auch in Minimalmedium in
Flüssigkultur vergleichbar zum Wildtyp. Setzt man die Stämme auf Festmedium
Einleitung
22
verschiedenen Stressfaktoren aus, so zeigt die PC-defiziente Mutante in
Anwesenheit des anionischen Detergenz SDS und bei erhöhter Temperatur einen
drastischen Wachstumsdefekt (Wessel et al., 2006). PC scheint somit in
A. tumefaciens für das normale Wachstum zwar keine wichtige Rolle zu spielen, ist
jedoch für die Stressadaption entscheidend.
Die PC-defiziente Mutante ist nach Infektion von K. diagremontiana-Blättern nicht in
der Lage, eine Tumorbildung zu induzieren, während der Wildtyp und die
Einzelmutanten deutliche Zellwucherungen hervorrufen (Abb. 9A). Dabei liegt jedoch
eine zeitlich leicht verzögerte und verminderte Tumorbildung nach Infektion mit der
ΔpmtA-Mutante vor. Dies ist auf eine reduzierte PC-Menge in den Membranen bei
Anzucht in Minimalmedium zurückzuführen (Wessel et al., 2006). Der Virulenzdefekt
der Doppelmutante wurde mit einem quantitativen Kartoffelassay als weitere,
unabhängige Methode bestätigt (Abb. 9B).
A) K. diagremontiana
B) Kartoffelscheiben
Abb. 9: Virulenzdefekt in Abwesenheit von Phosphatidylcholin. A) Dargestellt ist die Tumorbildung auf einem K. diagremontiana-Blatt nach Infektion mit A. tumefaciens Wildtyp und den PC-Biosynthesemutanten. Die Zellen wurden entsprechend ihrer optischen Dichte (A600) eingestellt, verdünnt und auf die verletzte Pflanzenoberfläche gegeben (modifiziert nach Wessel et al., 2006). B) Beispielhafte Darstellung zweier Kartoffelscheiben die entweder mit A. tumefaciens Wildtyp oder der ΔpmtA Δpcs-Mutante infiziert wurden. Angegeben ist die durchschnittliche Tumorbildung mit Standardfehler aus zwei unabhängigen Experimenten mit jeweils 120 Kartoffelscheiben pro Stamm (Klüsener, unveröffentlicht).
Der oben beschriebene Virulenzdefekt der ΔpmtA Δpcs-Mutante basiert auf der
Abwesenheit des TIVSS, welches die T-DNA und die VirE2-Proteine in die
Pflanzenzelle transferiert. Reportergenfusionen zeigen, dass die Expression der für
das TIVSS kodierenden Gene (virB/D) in der Mutante unter Virulenzbedingungen
nicht induziert wird. Die für die Induktion der virB/D-Genexpression notwendigen
Gene virA/G des Zwei-Komponentensystems sind in der ΔpmtA Δpcs-Mutante basal
Einleitung
23
exprimiert (Wessel et al., 2006). Ob die Proteine dieses Zwei-Komponentensystems
korrekt assembliert und funktionell sind, muss noch untersucht werden.
4. Zielsetzung
In A. tumefaciens wurden die für die PC-Biosynthese verantwortlichen Gene, pmtA
und pcs, erfolgreich identifiziert (Wessel et al., 2006). Im Mittelpunkt dieser Arbeit
stand die Charakterisierung der PC-Biosynthese in A. tumefaciens auf DNA-,
Protein- und Lipidebene. Dazu sollte die genetische Organisation der pmtA- und pcs-
Gene mittels RT-PCR und eine mögliche Regulation der Genexpression durch
Reportergenfusionen unter verschiedenen Bedingungen analysiert werden. Des
Weiteren sollten die pmtA- und pcs-Gene kloniert, in E. coli exprimiert und
charakterisiert werden. Eine mögliche Lokalisation der entstehenden PC-Moleküle in
der inneren und äußeren Membran sollte über eine Membrantrennung und
nachfolgender Dünnschichtchromatographie untersucht werden.
Für A. tumefaciens (Wessel et al., 2006), B. japonicum (Minder et al., 2001),
B. abortus (Comerci et al., 2006; Conde-Alvarez et al., 2006) und L. pneumophila
(Conover et al., 2008) konnte eine wichtige Rolle für PC in der Interaktion mit dem
eukaryotischen Wirt gezeigt werden. Jedoch sind die molekularen Ursachen oft noch
nicht geklärt und weitere mögliche Funktionen von PC in Bakterien völlig unbekannt.
Einen Schwerpunkt dieser Arbeit stellen vergleichende Transkriptom- und
Proteomanalysen von A. tumefaciens Wildtyp und PC-defizienter Mutante sowie
einer phänotypischen Charakterisierung der Mutante dar, um die globale
physiologische Relevanz von PC aufzuklären.
Neben der Charakterisierung und Bedeutung der PC-Biosynthese in A. tumefaciens
sollte in einem weiteren Projekt die Virulenz-induzierte Expression des kleinen
Hitzeschockproteins HspL untersucht und eine mögliche Funktion des Proteins in der
Virulenz identifiziert werden. Dazu sollte eine putative Rolle von HspL in der
Assemblierung des für den T-DNA Transfer essentiellen Typ IV-Sekretionssystems
analysiert werden.
PC biosynthesis in Agrobacterium tumefaciens
24
– B –
Expression and physiological relevance of Agrobacterium tumefaciens
phosphatidylcholine biosynthesis genes
Sonja Klüsener, Meriyem Aktas, Kai M. Thormann, Mirja Wessel and
Franz Narberhaus
JOURNAL OF BACTERIOLOGY, Jan. 2009, p. 365–374 Vol. 191, No. 10021-9193/09/$08.00�0 doi:10.1128/JB.01183-08Copyright © 2009, American Society for Microbiology. All Rights Reserved.
Expression and Physiological Relevance of Agrobacterium tumefaciensPhosphatidylcholine Biosynthesis Genes�
Sonja Klusener,1# Meriyem Aktas,1# Kai M. Thormann,1,2 Mirja Wessel,1 and Franz Narberhaus1*Microbial Biology, Ruhr-University Bochum, Bochum, Germany,1 and Department of Ecophysiology, Max Planck Institute for
Terrestrial Microbiology, Marburg, Germany2
Received 22 August 2008/Accepted 23 October 2008
Phosphatidylcholine (PC), or lecithin, is the major phospholipid in eukaryotic membranes, whereas only10% of all bacteria are predicted to synthesize PC. In Rhizobiaceae, including the phytopathogenic bacteriumAgrobacterium tumefaciens, PC is essential for the establishment of a successful host-microbe interaction. A.tumefaciens produces PC via two alternative pathways, the methylation pathway and the Pcs pathway. Theresponsible genes, pmtA (coding for a phospholipid N-methyltransferase) and pcs (coding for a PC synthase),are located on the circular chromosome of A. tumefaciens C58. Recombinant expression of pmtA and pcs inEscherichia coli revealed that the individual proteins carry out the annotated enzyme functions. Both genes anda putative ABC transporter operon downstream of PC are constitutively expressed in A. tumefaciens. Theamount of PC in A. tumefaciens membranes reaches around 23% of total membrane lipids. We show that PCis distributed in both the inner and outer membranes. Loss of PC results in reduced motility and increasedbiofilm formation, two processes known to be involved in virulence. Our work documents the critical impor-tance of membrane lipid homeostasis for diverse cellular processes in A. tumefaciens.
Phosphatidylcholine (PC), or lecithin, is the most-abundantphospholipid in eukaryotic membranes. Apart from its struc-tural function in membrane bilayers and lipoproteins, PC isinvolved in many signal transduction pathways (2). Meanwhile,PC has also been found in an increasing number of bacteria, inparticular in species that interact with eukaryotic hosts (27). Inbacteria, PC is synthesized either by the methylation pathwayor by the Pcs pathway (Fig. 1). In the first pathway, the pre-cursor phosphatidylethanolamine (PE) is methylated in threereactions by one or several phospholipid N-methyltransferases(Pmt enzymes) via the intermediates monomethylphosphati-dylethanolamine (MMPE) and dimethylphosphatidylethanol-amine (DMPE) to form PC. Each reaction step requires S-adenosylmethionine as the methyl donor. In the Pcs pathway,PC is produced via a direct condensation of choline and CDP-diacylglycerol. This reaction is catalyzed by the PC synthase(Pcs), an enzyme unique to prokaryotes (42).
The gram-negative alphaproteobacterium Agrobacterium tu-mefaciens is most commonly known for causing crown galldisease in plants. It synthesizes PC by using both the methyl-ation pathway and the Pcs pathway (Fig. 1) (21, 23). The latterpathway requires choline in the medium. During growth inminimal medium, only marginal amounts of PC were formedby the Pcs enzyme (45). In Brucella abortus, PC production viathe Pcs pathway has been reported to depend on the presenceof choline provided by the host (5). The plant symbiont Sino-rhizobium meliloti uses plant-exuded choline for PC biosynthe-sis (13). The finding that A. tumefaciens cannot produce cho-
line de novo (41) implies that there must be a choline uptakesystem to supply the Pcs pathway with choline. The only evi-dence for a choline uptake system so far is that A. tumefacienswas shown to take up radiolabeled choline from the cultivationmedium (41). The responsible transporter, however, remainselusive.
Several symbiotic and pathogenic bacteria depend on PC toestablish a successful host-microbe interaction. We demon-strated previously that an A. tumefaciens mutant lacking bothpredicted PC biosynthesis pathways did not produce any PCand was incapable of tumor formation on plant leaves (45).This virulence defect was caused by the absence of the mem-brane-spanning type IV secretion system, which delivers theoncogenic T-DNA from the tumor-inducing (Ti) plasmid andeffector proteins to plant cells (4, 47). In Bradyrhizobium ja-ponicum, the nitrogen-fixing symbiont of the soybean Glycinemax, PC is required to establish functional root nodules (31).PC also is necessary for full virulence of the human pathogensB. abortus (6) and Legionella pneumophila (7). Although thisstrongly suggests that PC is a critical determinant in host-microbe interactions, there still is little understanding re-garding the exact role that PC might play in these processes.Moreover, it is largely unclear whether the expression of PCbiosynthesis genes is regulated. The first hints that the ex-pression of PC biosynthesis genes might be subject to envi-ronmental control were obtained in B. japonicum. This or-ganism encodes a pmt multigene family comprised of pmtAand four additional pmt genes (18). Under normal condi-tions, only pmtA and pmtX1 are expressed. However, in apmtA mutant, pmtX3 and, in particular, pmtX4 are expressed(17, 18).
In this study we examined the expression and physiologicalrole of A. tumefaciens PC biosynthesis genes pmtA and pcs. Themost important findings are that PC produced by these en-zymes is distributed in the inner and outer membrane and that
* Corresponding author. Mailing address: Lehrstuhl fur Biologie derMikroorganismen, Fakultat fur Biologie und Biotechnologie, Ruhr-Universitat Bochum, D-44780 Bochum, Germany. Phone: 49 (234) 3223100. Fax: 49 (234) 32 14620. E-mail: franz.narberhaus@rub.de.
# S.K. and M.A. contributed equally to this study.� Published ahead of print on 31 October 2008.
365
at UN
IVE
RS
ITA
T B
OC
HU
M on F
ebruary 19, 2009 jb.asm
.orgD
ownloaded from
25
the absence of PC is associated with severe phenotypes inmotility and biofilm formation.
MATERIALS AND METHODS
Bacterial strains and growth conditions. All strains and plasmids used in thiswork are listed in Table 1. Oligonucleotides are listed in Table 2. Escherichia colicells were grown at 37°C in Luria-Bertani (LB) medium (38), supplemented withkanamycin, streptomycin, and/or spectinomycin at a final concentration of 50�g/ml if appropriate. E. coli DH5� was used as host for all cloning procedures.E. coli BL21(DE3), which contains the phage T7 polymerase gene under thecontrol of the lacUV5 promoter (43), served as the host for overproduction ofPmtA and Pcs from the corresponding pET24b-based expression plasmids. A.tumefaciens strain C58 (wild type) and its derivatives (�pmtA, �pcs, �pmtA �pcs,and �pmtA �abc mutants) were routinely grown at 30°C in YEB complex or ABminimal medium (pH 5.5, 1% [wt/vol] glucose) (40), supplemented with 100
�g/ml ampicillin, 2 �g/ml tetracycline, 50 �g/ml kanamycin, 100 �g/ml strepto-mycin, and/or 300 �g/ml spectinomycin if necessary. For �-galactosidase assays,choline (Sigma-Aldrich, Munchen, Germany) was added to the AB medium infinal concentrations of 0.1 mM, 0.5 mM, or 1 mM. In support of virulence geneinduction, Agrobacterium cells were precultivated to an optical density at 600 nm(OD600) of 0.2 at 30°C in liquid AB minimal medium (pH 5.5) with 1% (wt/vol)glucose prior to the addition of acetosyringone. Cells were further incubatedfor 16 to 20 h at 23°C. Acetosyringone was always used at a final concentra-tion of 0.1 mM.
Plasmid and mutant construction. Recombinant DNA work was carried outaccording to standard protocols (38). PCR-generated fragments of the promoterregions of pmtA, pcs, and abc (16, 48) were digested with KpnI and XhoI andligated into pAC01 (26) treated with the same enzymes to construct transcrip-tional fusions to the lacZ gene. For overproduction of the agrobacterial PmtAand Pcs in E. coli, the expression plasmids pET_PmtA and pET_Pcs wereconstructed. The pmtA and pcs genes were amplified by PCR. The fragmentswere cleaved with NdeI/SalI and NdeI/HindIII, respectively, and ligated intopET24b treated with the same enzymes. For overproduction of PmtA and Pcs inA. tumefaciens, the coding regions, including their own ribosome binding site,were amplified via PCR. The PCR products were digested with EcoRI/SalI andEcoRI/HindIII, respectively, and cloned into the corresponding sites of thevector pVS-BADNco (49), resulting in the plasmids pVS-PmtA and pVS-Pcs.The markerless abc1-4 deletion was introduced into the A. tumefaciens �pmtAstrain as described previously (45) by using the suicide vector pK19mobsacB. Theabc1-4 up- and downstream regions were amplified by PCR. The PCR product ofthe abc1-4 upstream region was digested with EcoRI/PstI and cloned into thevector pK19mobsacB, resulting in the plasmid pK19mobsacB_abc_up. The PCRproduct of the downstream region was digested with PstI/HindIII and cloned intopK19mobsacB_abc_up, resulting in the plasmid pK19mobsacB_abc_updo. Thecorrect nucleotide sequences of all plasmids constructed were confirmed byautomated sequencing. Plasmids were transferred into E. coli via transformationand into A. tumefaciens via electroporation.
Overproduction of PmtA and Pcs in E. coli. E. coli BL21(DE3) carryingpET_PmtA or pET_Pcs was cultivated in LB medium at 37°C until the OD600
FIG. 1. PC biosynthesis pathways in A. tumefaciens C58. As indi-cated by the question marks, the choline uptake system is unknown.SAM, S-adenosylmethionine; SAH, S-adenosylhomocysteine; PmtA,phospholipid N-methyltransferase; Pcs, phosphatidylcholine synthase.
TABLE 1. Strains and plasmids used in this study
Strain or plasmid Relevant characteristic(s)a Reference or source
StrainsEscherichia coli
DH5� Cloning host 19BL21(DE3) Expression host 43
Agrobacterium tumefaciensC58 Wild type C. Baron, Montreal,
CanadaC58 �pmtA Derivative of the wild type with deletion of the pmtA gene 45C58 �pcs Derivative of the wild type with deletion of the pcs gene 45C58 �pmtA �pcs Derivative of the wild type with deletion of the pmtA and pcs genes 45C58 �pmtA �abc Derivative of the wild type with deletion of the pmtA and abc1-4 genes This study
PlasmidspAC01 Tcr Apr; transcriptional lacZ fusion vector containing promoterless lacZ gene 26pET24b Kmr; high-copy His-tag expression vector Novagen, Darmstadt,
GermanypVS-BADNco Spr Str; A. tumefaciens expression vector 49pK19mobsacB Kmr; suicide vector 39pBO380 Tcr Apr; pmtA-lacZ fusion in pAC01 This studypBO377 Tcr Apr; pcs-lacZ fusion in pAC01 This studypBO1264 Tcr Apr; abc-lacZ fusion in pAC01 This studypBO801 Kmr; derivative of pET24b for overproduction of PmtA with a C-terminal His tag This studypBO803 Kmr; derivative of pET24b for overproduction of Pcs with a C-terminal His tag This studypBO813 Spr Str; derivative of pVS-BADNco carrying pmtA This studypBO814 Spr Str; derivative of pVS-BADNco carrying pcs This studypBO308 Kmr; derivative of pK19mobsacB carrying the upstream region of abc1-4 This studypBO309 Kmr; derivative of pK19mobsacB carrying the up- and downstream regions of
abc1-4This study
pLacTac-Gfp lacIq Ptac::gfp-mut2 Tcr repA oriVpVSI oriVp15A oriT K. Thormann, MarburgGermany
a Ap, ampicillin; Km, kanamycin; Sp, spectinomycin; St, streptomycin; Tc, tetracycline.
366 KLUSENER ET AL. J. BACTERIOL.
at UN
IVE
RS
ITA
T B
OC
HU
M on F
ebruary 19, 2009 jb.asm
.orgD
ownloaded from
26
reached a value between 0.5 and 0.8. Then, the synthesis of PmtA or Pcs wasinduced by the addition of isopropyl-�-D-thiogalactopyranoside (IPTG) to a finalconcentration of 0.4 mM and the cultures were incubated for another 2 h at 30°C.Subsequently, 1 ml culture was harvested by centrifugation. Cell pellets wereresuspended in 1� sodium dodecyl sulfate (SDS) loading buffer according to theOD600 (OD600 of 1 � 100 �l 1� SDS loading buffer) and boiled for 10 min. Tenmicroliters of each sample was separated on 12.5% SDS-polyacrylamide gels,and the proteins were stained with Coomassie blue.
Lipid analysis by TLC. The lipid composition of A. tumefaciens and E. colistrains was determined via thin-layer chromatography (TLC). Cells were culti-vated as mentioned above, harvested by centrifugation, washed with 500 �lwater, and resuspended in 100 �l water. The lipids were extracted according tothe method of Bligh and Dyer (1), separated by one-dimensional thin-layerchromatography using HPTLC silica gel 60 plates (Merck, Darmstadt, Ger-many), and stained with molybdenum blue spray reagent (Sigma-Aldrich) orCu2SO4 solution [300 mM copper(II)-sulfate-pentahydrate, 8.5% (vol/vol) phos-phoric acid]. In the case of the separated inner and outer membranes, the lipidextraction was started directly from the collected fractions. PE, MMPE, DMPE,and PC were used as phospholipid standards (Sigma-Aldrich), and n-propanol–propionate–chloroform–water (3:2:2:1) as the running solvent.
RT-PCR. A. tumefaciens strains were cultivated in YEB complex mediumuntil exponential phase. Total RNA was isolated by using a Micro-to-Midi totalRNA purification system (Invitrogen, Karlsruhe, Germany). The RNA was fur-ther treated with DNase I (amplification grade; Invitrogen) as specified by themanufacturer to remove contaminating chromosomal DNA. Reverse transcrip-tase (RT)-PCR and subsequent PCRs were performed according to the manu-facturer’s manual for the ThermoScript RT-PCR system (Invitrogen). The prim-
ers used for these experiments are listed in Table 2 (see also Fig. 3A and C). Thereaction products were separated by electrophoresis in 2% (wt/vol) agarose gelsand visualized by staining with ethidium bromide.
�-Galactosidase assays. The �-galactosidase activity of A. tumefaciens cellsgrown in liquid YEB complex medium or in AB minimal medium was measuredaccording to standard protocols (30). The plasmid pAC01 (26) containing thepromoterless lacZ gene was used as the negative control.
Separation of inner and outer membrane. Membrane separation was per-formed as described previously (12), with minor modifications. Four hundredmilliliters YEB cell culture was grown to an OD600 of 0.5 to 0.6 at 30°C andharvested by centrifugation at 10,000 � g, 4°C, for 10 min. The cells wereresuspended in 24 ml lysis buffer (50 mM Tris-HCl, pH 7.5, 20% [wt/vol] sucrose,0.2 M KCl, 0.2 mM dithiothreitol, 0.2 mg/ml DNase I, 0.2 mg/ml RNase A, 1 mMphenylmethylsulfonyl fluoride) and disrupted by two passes through a chilledFrench pressure cell at 16,000 lb/in2. The lysate was treated with 0.5 mg/mllysozyme for 1 h on ice and centrifuged at 10,000 � g, 4°C, for 20 min to removethe unbroken cells. The supernatant was centrifuged at 150,000 � g (SW40Ti),4°C, for 1 h in an ultracentrifuge to collect the membranes. The resultingmembrane pellet was carefully resuspended in 2 ml of 20% (wt/vol) sucrosecontaining 5 mM EDTA, pH 7.5, and 0.2 mM dithiothreitol. The resuspendedmembranes were centrifuged for 5 min at 16,000 � g to remove the insolublemembranes. The gradient was prepared by layering 7.5 ml 53% (wt/vol) sucroseover a cushion of 2.5 ml 70% (wt/vol) sucrose. Both sucrose solutions contained5 mM EDTA, pH 7.5. The membrane suspension was layered on the top of thegradient, and sucrose density gradient ultracentrifugation was carried out at100,000 � g (SW40Ti), 4°C, for 16 h. After ultracentrifugation, the separatedmembranes were fractionated in 500-�l aliquots to analyze the protein concen-
TABLE 2. Oligonucleotides used in this study
Purpose Oligonucleotide Sequence (5’33’)a
Transcriptional lacZ fusions pmtA_Fw_KpnI AAAAGGTACCGATTTTTCTGCTGGCGpmtA_Rv_XhoI AAAACTCGAGGATTCCCCTCGGATGpcs_Fw_KpnI AAAAGGTACCCGGTTGTGTCGAGGpcs_Rv_XhoI AAAACTCGAGGGTCAGCGCCTGCabc_Fw_KpnI AAAAGGTACCCGCTTAATCTCCTCGabc_Rv_XhoI AAAACTCGAGGAGTTCCCTCTCTC
RT-PCR pmtA_K1 (P1) GCAACGGCTTGAACAAAAATpmtA_K2 (P2) GAATATTGCGCACGATGAAApmtA_Kontr_fw (N1) GCTGATCGACCGGACGCAGpmtA_RT_PE3 (N2) ATTTTTGTTCAAGCCGTTGCpmtA_do_fw_BamHI (A) CGCGGATCCGGCGCAATTGTGGACCTATpmtA_Kontr_rv (B) GCCCAGGGAAGGCGGCTCGpcs_K1 (P3) GCGCCTTTTCCGTTCATATCpcs_K2 (P4) GTTTCGAAATGCATCAGCAGSig219 (N3) TTCAGGCTCAAGGTTTCCAGpcs_RT_PE (N4) GTTTTTCCATCCGCCCTGTtcapcs_1 (C) GCGAGCGGCATCTATCTTTApcs_2 (D) GAAATTGCCGAACAGCTTG
Overproduction of PmtA and Pcsin E. coli
pmtA_Nde_fw GGAATTCCATATGGCACTCAACCTGAAGCAAC
pmtA_Sal_Xho_rv CCGGTCGACTCACTCGAGGGCCCGCGTATAGGTCCACAApcs_Nde_fw GGAATTCCATATGGCCGCATACCGGAATGAApcs_Hind_Xho_rv CCCAAGCTTTCACTCGAGTTTCGCGCCGAGCGAGGGGAA
Overproduction of PmtA and Pcsin A. tumefaciens
pmtA_EcoR_fw CCGGAATTCTGACAATGACCAATACATCCGAG
pmtA_Sal_rv GCAGGTCGACTCAGGCCCGCGTATAGGTCCACpcs_EcoR_fw2 CCGGAATTCTGAGCGGATGGAAAAACGGGGCGpcs_Hind_rv CCCAAGCTTTTATTTCGCGCCGAGCGAGGGG
Construction of a markerlessabc1–4 deletion mutant in A.tumefaciens °pmtA
abc_up_fw_EcoRI CGGAATTCCGTCACCTATGTGCTCTTGCCCGCC
abc_up_rv_PstI AACTGCAGGAACAGCTTGGTCAGCTTGCGabc_do_fw_PstI AACTGCAGGCTCGGCCTCAACTTCTACGTTabc_do_rv_HindIII CCCAAGCTTGCCTTGATCAACGAACCAACGG
a Restriction sites in the oligonucleotides are underlined.
VOL. 191, 2009 PC BIOSYNTHESIS IN AGROBACTERIUM TUMEFACIENS 367
at UN
IVE
RS
ITA
T B
OC
HU
M on F
ebruary 19, 2009 jb.asm
.orgD
ownloaded from
27
tration, the NADH activity, and the phospholipid pattern. The protein concen-tration was determined by using a Bradford assay (Bio-Rad Laboratories GmbH,Munchen, Germany). The NADH oxidase activity was detected by the method ofOsborn et al. (32).
Motility assay. Motility assays were carried out on AB minimal mediumsolidified with 0.3% (wt/vol) agar (Difco, Lawrence, KS). A single colony from aYEB agar plate was inoculated onto the surface of the motility plates. The plateswere examined after 48 h of incubation at 23°C. To check whether calcium and/ormagnesium influence motility, MgSO4 was added in final concentrations of 0.24mM, 1.2 mM, or 12 mM and CaCl2 was added in final concentrations of 0.014mM, 0.07 mM, or 0.7 mM to AB medium.
SDS-PAGE and Western blotting. Wild-type and mutant A. tumefaciens cellswere cultivated as mentioned above, and 1 ml of each culture was harvested andresuspended in 1� SDS loading buffer in relation to the OD600 (OD600 of 1 �100 �l 1� SDS loading buffer). The samples were incubated for 10 min at 95°C,separated by 12.5% SDS-polyacrylamide gel electrophoresis (PAGE), and blot-ted onto polyvinylidene difluoride membranes (Bio-Rad). Detection was per-formed with a chemiluminescence-based system (Pierce Biotechnology, Rock-ford, IL) using A. tumefaciens flagellum protein-specific antiserum (1:30,000).
Flow cell biofilms. Biofilms of A. tumefaciens strains carrying plasmid-encodedgreen fluorescent protein (GFP) were cultivated at 30°C in three-channel flowcells with individual channel dimensions of 1 by 4 by 40 mm. Each flow chamberwas prepared by gluing a boron-silicate glass microscope coverslip, which servedas a substratum for microbial attachment, onto the flow chamber with siliconeand leaving it to dry for 24 h at room temperature prior to use. The assembly ofthe flow system was carried out essentially as described earlier (44).
The appropriate A. tumefaciens strains were cultivated until reaching theexponential or the stationary growth phase. The OD600 was then adjusted to 0.01in AB minimal medium with 0.5% (vol/vol) glycerol. One milliliter of the dilutedcell suspension was injected into each channel after the flow of medium wasarrested, and the chambers were turned upside down to facilitate initial attach-ment. After 30 min of incubation at 30°C, the flow cells were inverted, and theflow of medium was started at a constant rate of 75 �l/min per channel, using aWatson-Marlow Bredel 205S peristaltic pump (Cornwall, United Kingdom). Allbiofilm characterizations were conducted in triplicate in at least two independentexperiments.
Sixty minutes prior to microscopy, gfp expression from plasmid pLacTac-Gfpwas induced by the addition of IPTG (1 mM). Microscopic visualization ofbiofilms was carried out using an inverted Zeiss LSM510 confocal laser scanningmicroscope (Carl Zeiss, Jena, Germany) equipped with the following objectives:10�/0.3 W Plan-Neofluar, 20�/0.5 W Achroplan, and 40�/1.2 W C-Apochro-mat. For quantitative analysis, the image data were further processed using theIMARIS software package (Bitplane AG, Zurich, Switzerland) and Adobe Pho-toshop.
RESULTS AND DISCUSSION
Functional analysis of pmtA and pcs gene products. We havepreviously shown that an A. tumefaciens �pmtA �pcs mutant isno longer able to synthesize PC (45). To provide conclusiveevidence that PmtA and Pcs act as phospholipid N-methyl-transferase and PC synthase, respectively, we cloned pmtA andpcs into the expression vector pET24b, resulting in plasmidspET_PmtA and pET_Pcs. The analysis of crude protein ex-tracts from E. coli cells containing the first plasmid by SDS-PAGE revealed an overproduced protein of 22 kDa (Fig. 2A),which is in close agreement with the calculated mass of theagrobacterial pmtA gene product. In contrast, overexpressionof Pcs (calculated mass of 29 kDa), which is thought to be amembrane protein, was not observed (Fig. 2A). Either Pcs wasnot synthesized at all or it was synthesized in amounts belowthe detection limit.
To obtain proof for the biochemical activity of both recom-binant proteins, we compared the membrane lipid contents ofE. coli BL21(DE3) cells carrying pET_PmtA, pET_Pcs, or theempty vector pET24b. E. coli membranes are known to containthe lipids PE, phosphatidylglycerol, and cardiolipin (8), whichmigrate as a single spot in one-dimensional TLC (Fig. 2B). As
expected, E. coli BL21(DE3)/pET24b did not produce PC ormethylated intermediates. The membranes of E. coli cells ex-pressing pmtA consist of PE, MMPE, DMPE, and PC (Fig.2B). This is consistent with the results described for a Sinorhi-zobium meliloti pmtA gene expressed in E. coli BL21(DE3)(31). Only PC and no methylated intermediates were producedwhen the Rhodobacter sphaeroides pmtA gene was expressed inE. coli (31). The expression of the B. japonicum pmtA gene inE. coli BL21(DE3) led predominantly to the formation ofMMPE, with some DMPE but only marginal amounts of PC(18). Apparently, different Pmt enzymes possess different sub-strate and product specificities which cannot be predictedsolely on the basis of their primary sequence.
As the Pcs enzyme uses choline directly to form PC, the E.coli strain containing pET_Pcs produces PE and PC but nomono- or dimethylated intermediates (Fig. 2B). It is notablethat significant amounts of PC accumulated in this strain al-though the Pcs protein was only poorly expressed (Fig. 2A).Our results demonstrate that both recombinant proteins areactive when expressed in E. coli and are probably responsiblefor the biosynthesis of PC in A. tumefaciens without the re-quirement of additional proteins.
FIG. 2. Activity of agrobacterial PmtA and Pcs after expression inE. coli. (A) Detection of agrobacterial phospholipid N-methyltrans-ferase (PmtA) and PC synthase (Pcs) in crude extracts of E. coli cellsvia SDS-PAGE. E. coli BL21(DE3) cells with pET_PmtA, pET_Pcs, orpET24b were cultivated in LB complex medium, and protein expres-sion was induced with IPTG. Protein bands were visualized by Coo-massie blue staining. M, BenchMark protein standard (Invitrogen). �,present; �, absent. (B) Lipid formation after expression of agrobac-terial PmtA and Pcs in E. coli BL21(DE3). Lipids of BL21(DE3)derivatives were extracted and separated by one-dimensional TLC.Phospholipids were specifically stained with molybdenum blue sprayand compared to phospholipid standards in lane M (PE, MMPE,DMPE, and PC).
368 KLUSENER ET AL. J. BACTERIOL.
at UN
IVE
RS
ITA
T B
OC
HU
M on F
ebruary 19, 2009 jb.asm
.orgD
ownloaded from
28
Genetic organization of the Agrobacterium pmtA and pcsgene regions. The pmtA gene is flanked by two open readingframes, namely dnaN and atu0299, which are oriented in thesame direction as pmtA (Fig. 3A). The pmtA gene is separatedfrom dnaN and atu0299 by 195 bp and 95 bp, respectively. Toanalyze whether pmtA is cotranscribed with dnaN and/oratu0299, RT-PCR experiments were carried out. Total RNAwas isolated from a wild-type A. tumefaciens culture grown inYEB complex medium. Primers P2, N2, and B were used forRT reactions to produce cDNA. Primer pairs P1 and P2, N1and N2, and A and B were applied for PCR amplification (Fig.3A). As the positive control, PCRs were carried out with chro-mosomal DNA as template. Reaction mixtures without RTserved as the control for the absence of chromosomal DNA inour RNA preparations and did not produce amplificationproducts (Fig. 3B). The expected product of 591 bp featuringan internal pmtA fragment appeared only in the presence ofRT, showing that the gene was expressed. The presence of a258-bp PCR product with primer combination A and B sug-gests cotranscription of pmtA and atu0299. Since atu0299, pyrF,and atu0297 are separated by only 4 and 19 bp, respectively,it seems reasonable that these genes are cotranscribed.Thus, pmtA-atu0299-pyrF-atu0297 might form a tetracis-tronic operon. No PCR product was detectable with primerpair N1 by N2, indicating that dnaN and pmtA belong to dif-ferent transcription units (Fig. 3B).
The second PC biosynthesis gene, pcs, is located 212 bpdownstream of ubiH, encoding a predicted 2-octaprenyl-6-methoxyphenol hydroxylase which is transcribed in the oppo-site direction (Fig. 3C). Two hundred one base pairs furtherdownstream of pcs are four genes oriented in the same direc-tion. The gene products are annotated as an ABC transporter
with unknown substrate. We tested whether the genes abc1 toabc4 might form an operon with pcs. The primer pair P3 andP4, covering an internal DNA fragment of pcs, yielded theexpected product of 586 bp (Fig. 3D), demonstrating that thepcs gene is expressed under the tested conditions. In contrast,the primer set N3 and N4, designed to amplify a DNA frag-ment overlapping the borders of ubiH and pcs, did not producea PCR product, as expected since the two genes are oriented inopposite directions. The primer pair C and D, designed toamplify a DNA fragment overlapping the gene borders of pcsand abc1, resulted in a PCR product of 342 bp (Fig. 3D).Hence, it appears that pcs and the abc1-4 genes form anoperon.
Constitutive expression of A. tumefaciens PC biosynthesisgenes. To further analyze the expression of pmtA and pcs,transcriptional lacZ fusions were constructed (Fig. 3A and C).In addition, an abc1-lacZ fusion was created to test whetherabc expression depends exclusively on the pcs promoter (Fig.3C). The resulting reporter gene plasmids, pBO380 (pmtA-lacZ), pBO377 (pcs-lacZ), and pBO1264 (abc1-lacZ), wereelectroporated into wild-type A. tumefaciens. To analyze if thepresence or absence of PC influences the expression of PCbiosynthesis genes and/or abc1, the lacZ fusions were alsointroduced into existing PC biosynthesis mutants. It is knownthat the two single deletion �pmtA and �pcs mutants are ableto synthesize PC through the remaining pathway, whereas thedouble deletion �pmtA �pcs mutant is entirely PC deficient(45). All reporter strains were grown in YEB complex mediumand in AB minimal medium at 30°C. The PC biosynthesisgenes pmtA and pcs, as well as abc1, were clearly expressedunder all conditions tested. The expression levels were three-to fourfold higher in minimal medium (data not shown). Ex-
FIG. 3. RT-PCR of the pmtA and pcs gene regions. The PCR strategy is outlined in panels A and C. The positions of primers used for thereverse transcription and PCRs are given below the corresponding gene regions. Putative promoters and constructed lacZ fusions are indicated.dnaN encodes �-chain of DNA polymerase III; pmtA encodes phospholipid N-methyltransferase; pyrF encodes orotidine 5-monophosphatedecarboxylase; atu0299, atu0297, and atu0296 are hypothetical open reading frames; ubiH encodes 2-octaprenyl-6-methoxyphenol hydroxylase; pcsencodes PC synthase; abc1 encodes a nucleotide binding ABC transporter; abc2 and abc3 encode membrane-spanning proteins forming an ABCtransporter; and abc4 encodes a lipoprotein ABC transporter. In panels B and D, the results of RT-PCRs with RNA from wild-type A. tumefaciensC58 cells are presented. The primer pairs used and lengths of PCR products are indicated. c, PCR products using chromosomal DNA as template; �,standard RT-PCR; �, negative control in which no RT had been added to the reaction mixture.
VOL. 191, 2009 PC BIOSYNTHESIS IN AGROBACTERIUM TUMEFACIENS 369
at UN
IVE
RS
ITA
T B
OC
HU
M on F
ebruary 19, 2009 jb.asm
.orgD
ownloaded from
29
pression was not altered in the PC biosynthesis mutants (Fig.4), indicating that pmtA, pcs, and abc expression is indepen-dent of the presence of PC, MMPE, and DMPE. The signifi-cant �-galactosidase activity of the abc1-lacZ fusion demon-strates that the abc1-4 genes possess their own promoter inaddition to the one upstream of pcs (Fig. 3D). In all cases, thelevel of expression of pmtA was higher than that of pcs (Fig. 4).This might be taken as an additional line of support for ourprevious assumption that the methylation pathway is morerelevant than the Pcs pathway in A. tumefaciens. Stronger ev-idence for this conclusion derives from the following findings.Only traces of PC were detectable after growth in AB minimalmedium in the presence of choline (45), indicating that thestill-existing Pcs pathway cannot fully compensate for themethylation pathway. In addition, Kalanchoe leaves infectedwith the �pmtA mutant showed decreased tumor formation(45). The significance of the alternative PC biosynthesis path-ways in prokaryotes differs. In contrast to A. tumefaciens, thePcs pathway is of predominant importance in L. pneumophila(7) and Pseudomonas aeruginosa (46).
In S. meliloti, choline is taken up by an ABC transportsystem (14). Given the genetic organization of the abc1-4 genesimmediately downstream of the pcs gene (Fig. 3C), we specu-lated that choline might be a substrate of this putative ABCtransporter. In bacteria, the expression and activity of ABCtransporters is often regulated via their substrates. The expres-sion of the fructose uptake ABC transporter in S. meliloti isinduced in the presence of this sugar (25). Likewise, therhamnose uptake system in Rhizobium leguminosarum is con-trolled by rhamnose (36). To check if the expression of the abcoperon is choline dependent, choline was added to the ABminimal medium at a final concentration of 0.1 mM. ThepmtA-lacZ fusion was considered a negative control since theexpression of the methylation pathway should not be depen-dent on choline. The expression of all three genes (abc1, pcs,and pmtA) was not altered in the presence of 0.1 mM choline
(Fig. 4) or 0.5 mM or 1 mM choline (data not shown), indi-cating that the expression of the Pcs pathway and the putativeABC transporter is choline independent.
We have previously shown that the deletion of pmtA and pcscauses a drastic virulence defect (45). Genes involved in cho-line uptake and metabolism in S. meliloti are highly expressedunder symbiotic conditions (14, 28). To further inspect the linkbetween virulence and PC biosynthesis in A. tumefaciens, wetested if virulence conditions change the expression of the PCbiosynthesis genes. The addition of 0.1 mM acetosyringone, anartificial virulence gene inducer, to the AB minimal mediumdid not affect the expression of pmtA, pcs, and abc1 (data notshown). We conclude that all three genes are constitutivelyexpressed and not regulated by (i) the PC content, (ii) theavailability of choline, and (iii) the presence of acetosyringone.
The abc1-4 genes do not encode a critical choline uptakesystem. PC formation in an A. tumefaciens pmtA mutant de-pends on the Pcs pathway and the presence of choline in thegrowth medium (45). To address whether the predicted ABCtransporter encoded downstream of the pcs gene is responsiblefor choline uptake, we monitored PC biosynthesis in a pmtAabc1-4 double mutant grown in YEB complex medium by TLCanalysis. As expected, wild-type A. tumefaciens membranesconsist of PE, MMPE, DMPE, and PC, whereas a �pmtAmutant produces only PC but not the intermediates MMPEand DMPE (Fig. 5). The �pmtA �pcs mutant is no longer ableto synthesize PC (45). Wild-type-like amounts of PC in thepmtA abc1-4 deletion strain (Fig. 5) indicate that the Pcs path-way is efficiently supplied with its substrate choline. We con-clude that the abc1-4 genes are not responsible for cholineuptake or that additional transporters compensate for the lackof the ABC transporter. Residual choline uptake activity hasalso been reported in the absence of the high-affinity ChoXsystem in S. meliloti (14). Bacillus subtilis contains two closelyrelated ABC transport systems for the uptake of choline asprecursor for the osmoprotectant glycine betaine (22). Allthese findings suggest that bacteria might be equipped withalternative routes for choline uptake. Based on its genomesequence, A. tumefaciens is predicted to contain 667 compo-nents of ABC transporters (16, 48). There clearly is potentialfor alternative choline transport systems.
Membrane localization of PC. The results shown in Fig. 5and previous TLC analyses (23, 43) suggested that the relativePC content in A. tumefaciens membranes amounts to approx-
FIG. 4. �-Galactosidase activities of plasmid-encoded transcrip-tional lacZ fusions in wild-type A. tumefaciens C58 and PC biosynthesismutants. Cells were grown in AB minimal medium in the absence orpresence of 0.1 mM choline at 30°C. The error bars indicate standarddeviations of the results from three independent assays. The plasmidpAC01 containing the promoterless lacZ gene was used as negativecontrol, and the background activity was below 3 Miller units (MU). 1,wild-type; 2, �pmtA; 3, �pcs; 4, �pmtA �pcs.
FIG. 5. PC formation in the A. tumefaciens abc1-4 mutant. A. tu-mefaciens wild-type, �pmtA, �pmtA �pcs, and �pmtA �abc cells weregrown in YEB complex medium at 30°C. Lipids were extracted, sep-arated by one-dimensional TLC, and visualized by Cu2SO4 staining.
370 KLUSENER ET AL. J. BACTERIOL.
at UN
IVE
RS
ITA
T B
OC
HU
M on F
ebruary 19, 2009 jb.asm
.orgD
ownloaded from
30
imately 25%. The quantification of radiolabeled phospholipidsrevealed that PC levels indeed reach 23% of total phospholip-ids (data not shown). Since it was unknown whether PC islocalized in the inner, the outer, or in both membranes of A.tumefaciens, we isolated membranes of wild-type A. tumefa-ciens and subsequently separated the inner and outer mem-branes by sucrose density gradient ultracentrifugation (Fig.6A). As reported for R. leguminosarum (12), two major lipidfractions were visible as a diffuse cream-colored upper bandand a distinct white lower band. Measuring the protein con-centration of the individual fractions revealed two major peakswith maxima in fractions 3 to 7 and 19 to 21 (Fig. 6B). Both
peaks contain a clearly distinguishable collection of proteinsspanning the entire molecular-mass range (Fig. 6C). Fractionsfrom the inner and outer membranes were clearly distinguish-able. To assign the peaks to a membrane compartment,NADH oxidase activity, which is a marker enzyme for the innermembrane, was determined. As expected, the NADH oxidaseactivity was predominantly found in fractions 3 to 7 (data notshown).
To determine the distribution of PC, membrane lipids frompooled fractions 3 to 7 and 19 to 21 were analyzed by TLC. Thepooled inner membrane fraction mainly consists of PE and PC,with little MMPE and only traces of DMPE (Fig. 6D). Theouter membrane is also composed of PE, MMPE, and PC.DMPE was not visible, probably because the amount was be-low the detection limit. We conclude that A. tumefaciens con-tains PC in both the inner and outer membrane.
PC and PE are also the most-abundant phospholipids in thepathogens L. pneumophila (20) and P. aeruginosa (46), and ithas been shown that they are localized in both membranecompartments. The presence of PC is critical for adaptation todifferent environmental conditions in these organisms. Theloss of PC in L. pneumophila led to reduced bacterial bindingto macrophages during the infection process (7), and PC seemsto be important for P. aeruginosa in specific responses to stress(46). An A. tumefaciens PC-deficient mutant showed sensitivitytoward elevated temperature and SDS (45). The presence ofPC in the outer membrane, which is in direct contact with theenvironment, might be important for coping with stressful con-ditions.
PC-deficient mutants are impaired in motility. Since thelipid composition impacts bacterial fitness, we hypothesizedthat PC might be of particular importance for biological pro-cesses mediated through the membranes. The presence of PCis required for the induction of the type IV secretion system inA. tumefaciens (45). Another process that relies on a complexmembrane-spanning system is bacterial motility. Swimmingmotility in A. tumefaciens is mediated by flagella which aretypically localized as a small tuft at or around a single cell pole(3). To investigate the motility phenotype, the wild-type strainand the single deletion �pmtA and �pcs mutants, as well as thePC-deficient mutant (�pmtA �pcs), were plated on soft agarmedium. Consistent with previous reports (29), A. tumefaciensis highly motile, as indicated by the formation of large concen-tric swim rings (Fig. 7A). Both single mutants showed swim-ming behavior comparable to that of the wild-type (data notshown). In contrast, the PC-deficient mutant was severely im-paired in motility. The motility defect was partially restoredwhen the �pmtA �pcs mutant was complemented with eitherpmtA or pcs expressed from low-copy-number vector pVS_P-mtA or pVS_Pcs. It is known that many members of the Rhi-zobiaceae are nonmotile in medium lacking divalent cationsbut retain good motility in medium containing calcium, mag-nesium, barium, or strontium (37). Therefore, we checkedwhether calcium and/or magnesium could rescue the motilitydefect of the double mutant and analyzed its motility in theabsence or presence of different concentrations of both ions.Neither the addition of MgSo4 nor of CaCl2 restored the mo-tility defect (data not shown). Motility was also inspected mi-croscopically. The wild type showed normal motility, whereas
FIG. 6. Localization of PC in membranes of wild-type A. tumefa-ciens C58. (A) Schematic representation of bands observed after cen-trifugation. (B) Separation of agrobacterial inner and outer membraneby discontinuous sucrose density gradient centrifugation. Fractionswere collected as 500-�l aliquots from the top of the sucrose gradient.Protein concentrations (F) and sugar densities (Œ) of the gradientfractions are shown. (C) Fractions 5 and 6 (IM), fraction 12, andfractions 20 and 21 (OM) were analyzed by SDS-PAGE and visualizedwith Coomassie blue staining. (D) Lipids were extracted, and fractions3 to 7 and 19 to 21 were pooled, followed by one-dimensional TLCanalysis and molybdenum blue staining. Barely detectable lipids aremarked with arrows. PE, MMPE, DMPE, and PC were used as stan-dards. M, phospholipid standard; IM, inner membrane; OM, outermembrane.
VOL. 191, 2009 PC BIOSYNTHESIS IN AGROBACTERIUM TUMEFACIENS 371
at UN
IVE
RS
ITA
T B
OC
HU
M on F
ebruary 19, 2009 jb.asm
.orgD
ownloaded from
31
the �pmtA �pcs mutant showed uncoordinated circular move-ment (data not shown).
A L. pneumophila PC-deficient mutant has also been shownto be impaired in motility (7). This phenotype was due to thelack of flagellum proteins. Hence, we examined whether theseproteins are present in the agrobacterial PC-deficient mutant.Total protein extracts of the wild type and the �pmtA �pcsmutant were analyzed by Western blotting using a flagellumprotein-specific antibody raised against flagellar proteins of A.tumefaciens (B. Scharf, personal communication). In contrastto the L. pneumophila PC mutant, the flagellar proteins FlaA(modified by glycosylation) (11), FlaB, FlaC, and FlaD weredetected in the A. tumefaciens PC-deficient mutant (Fig. 7B),indicating that reduced motility is not caused by the absence ofthe flagellar proteins. The presence of flagellum filaments wasconfirmed by silver nitrate staining of intact cells (data notshown).
PC influences A. tumefaciens biofilm formation. A. tumefa-ciens forms complex biofilms on abiotic surfaces and planttissues, with an equilibrium between the sessile and motilelifestyle (10). When microbes directly encounter a surface, the
outer membrane represents the interface between cell andsubstratum. It has been proposed that systems presumablyinvolved in the cell envelope stress response, such as the Cpx orRcs systems of E. coli, are participating in the transition fromthe planktonic to the attached life style (15, 34). In addition, ithas been shown that motility often correlates with the coloni-zation of surfaces. An aflagellate (Fla�) E. coli mutant is de-ficient in biofilm formation (35), and several pseudomonadsshow biofilm formation and architecture dependent on flagella,as well as type IV pili (24, 33). Therefore, we hypothesized thatchanges in the composition of the outer membrane might alsoinfluence the ability of A. tumefaciens to form microbial com-munities. To investigate biofilm formation in the absence ofPC, we compared the wild-type and PC-deficient mutant(�pmtA �pcs) strains in a hydrodynamic flow chamber systemin which the attached cells were constantly provided with freshAB minimal medium. After 24 h, the PC-deficient mutantformed thicker and denser communities than the wild type,with numerous large towering structures (Fig. 8A). The differ-ence became even more obvious after 48 h. In addition to theenhanced structure formation, the mutant strains displayedalmost-complete surface coverage, while the wild-type commu-nities exhibited more-unsettled surface areas (Fig. 8B). Exceptfor the increase in height and biomass, no alterations occurredin the basic architecture of the structures formed by the twostrains. Cells harvested during the exponential and stationarygrowth phases showed similar behavior in biofilm formation. Insummary, it appears that in the PC-deficient mutant, the equi-librium between the sessile and motile cells is shifted moretoward the side of attachment, favoring the accumulation oflarger amounts of biomass. Consistent with our results, a sim-ilar phenotype was described by Merritt et al. (29) for a non-motile agrobacterial mutant (�flgE). The biofilms formed bythe �flgE mutant showed increases in biomass, surface cover-age, and average structure height compared to those of thewild type. Interestingly, the �flgE mutant biofilms grown instatic culture were highly reduced relative to those of the wildtype (29). In preliminary experiments, we detected a similareffect for the PC-deficient mutant when grown in static culture(data not shown). Several reasons might account for the bio-
FIG. 7. Motility assay of an agrobacterial PC-deficient mutant andcomplemented strains. (A) Motility of the wild type, the PC-deficientmutant (�pmtA �pcs), and the mutant complemented with eitherpVS_pmtA or pVS_pcs was assayed on AB minimal medium platecontaining 0.3% (wt/vol) agar. (B) Cell lysates of wild type and PC-deficient mutant were analyzed by SDS-PAGE and Western blottingusing a flagellin antibody (1:30,000).
FIG. 8. Biofilm formation of a PC-deficient mutant of A. tumefa-ciens C58. Flow cell biofilms of wild-type and mutant (�pmtA �pcs)expressing GFP were grown in AB minimal medium at 30°C. Micro-scopic visualization of biofilms was carried out using an inverted ZeissLSM510 confocal laser-scanning microscope equipped with 20�/0.5 WAchroplan objective. Image data were obtained after 24 h (A) and 48 h(B) and further processed using the IMARIS software package (Bit-plane AG, Zurich, Switzerland) and Adobe Photoshop.
372 KLUSENER ET AL. J. BACTERIOL.
at UN
IVE
RS
ITA
T B
OC
HU
M on F
ebruary 19, 2009 jb.asm
.orgD
ownloaded from
32
film phenotype of the �pmtA �pcs mutant. First, it might beattributed to a motility defect similar to that of the �flgEmutant. Second, changes in the envelope stress response elic-ited by alterations in the membrane composition might play arole. Third, it is possible that the altered membrane composi-tion affects biofilm matrix composition. Exopolysaccharidesand cellulose are known to influence the attachment and bio-film formation of A. tumefaciens (9).
Although we are now beginning to appreciate the biologicalsignificance of bacterial PC biosynthesis, we are far from un-derstanding the molecular details of how PC influences bacte-rial physiology and host-microbe interactions. We revealedthat PC in the membrane of A. tumefaciens plays a critical rolein seemingly diverse processes, such as stress response, motil-ity, social behavior, and virulence. Further studies will be di-rected toward revealing the underlying mechanisms of PC ac-tion in bacterial membranes.
ACKNOWLEDGMENTS
We are grateful to Ehr-Min Lai for plasmid pAC01, Yun-long Tsaifor advice on membrane separation, and Birgit Scharf for antiflagellasera. We thank Christiane Fritz for excellent technical assistance,Stephanie Hacker for quantitative phospholipid analysis, RobbinStantscheff for mutant construction, and Bernd Masepohl for helpfulcomments on the manuscript.
The work was in part supported by a grant from the German Re-search Foundation (DFG; SFB 480) to F.N. and a fellowship from thePromotionskolleg of the Ruhr-University Bochum to M.A.
REFERENCES
1. Bligh, E. G., and W. J. Dyer. 1959. A rapid method of total lipid extractionand purification. Can. J. Biochem. Physiol. 37:911–917.
2. Canty, D. J., and S. H. Zeisel. 1994. Lecithin and choline in human healthand disease. Nutr. Rev. 52:327–339.
3. Chesnokova, O., J. B. Coutinho, I. H. Khan, M. S. Mikhail, and C. I. Kado.1997. Characterization of flagella genes of Agrobacterium tumefaciens, andthe effect of a bald strain on virulence. Mol. Microbiol. 23:579–590.
4. Christie, P. J., and E. Cascales. 2005. Structural and dynamic properties ofbacterial type IV secretion systems. Mol. Membr. Biol. 22:51–61.
5. Comerci, D. J., S. Altabe, D. de Mendoza, and R. A. Ugalde. 2006. Brucellaabortus synthesizes phosphatidylcholine from choline provided by the host. J.Bacteriol. 188:1929–1934.
6. Conde-Alvarez, R., M. J. Grillo, S. P. Salcedo, M. J. de Miguel, E. Fugier,J. P. Gorvel, I. Moriyon, and M. Iriarte. 2006. Synthesis of phosphatidyl-choline, a typical eukaryotic phospholipid, is necessary for full virulence ofthe intracellular bacterial parasite Brucella abortus. Cell Microbiol. 8:1322–1335.
7. Conover, G. M., F. Martínez-Morales, M. I. Heidtman, Z. Q. Luo, M. Tang,C. Chen, O. Geiger, and R. R. Isberg. 2008. Phosphatidylcholine synthesis isrequired for optimal function of Legionella pneumophila virulence determi-nants. Cell. Microbiol. 10:514–528.
8. Cronan, J. E. 2003. Bacterial membrane lipids: where do we stand? Annu.Rev. Microbiol. 57:203–224.
9. Danhorn, T., and C. Fuqua. 2007. Biofilm formation by plant-associatedbacteria. Annu. Rev. Microbiol. 61:401–422.
10. Danhorn, T., M. Hentzer, M. Givskov, M. R. Parsek, and C. Fuqua. 2004.Phosphorus limitation enhances biofilm formation of the plant pathogenAgrobacterium tumefaciens through the PhoR-PhoB regulatory system. J.Bacteriol. 186:4492–4501.
11. Deakin, W. J., V. E. Parker, E. L. Wright, K. J. Ashcroft, G. J. Loake, andC. H. Shaw. 1999. Agrobacterium tumefaciens possesses a fourth flagelin genelocated in a large gene cluster concerned with flagellar structure, assemblyand motility. Microbiology 145:1397–1407.
12. de Maagd, R. A., and B. Lugtenberg. 1986. Fractionation of Rhizobiumleguminosarum cells into outer membrane, cytoplasmic membrane, periplas-mic, and cytoplasmic components. J. Bacteriol. 167:1083–1085.
13. de Rudder, K. E., C. Sohlenkamp, and O. Geiger. 1999. Plant-exuded cholineis used for rhizobial membrane lipid biosynthesis by phosphatidylcholinesynthase. J. Biol. Chem. 274:20011–20016.
14. Dupont, L., I. Garcia, M. C. Poggi, G. Alloing, K. Mandon, and D. LeRudulier. 2004. The Sinorhizobium meliloti ABC transporter Cho is highlyspecific for choline and expressed in bacteroids from Medicago sativa nod-ules. J. Bacteriol. 186:5988–5996.
15. Ferrieres, L., and D. J. Clarke. 2003. The RcsC sensor kinase is required fornormal biofilm formation in Escherichia coli K-12 and controls the expres-sion of a regulon in response to growth on a solid surface. Mol. Microbiol.50:1665–1682.
16. Goodner, B., G. Hinkle, S. Gattung, N. Miller, M. Blanchard, B. Qurollo,B. S. Goldman, Y. Cao, M. Askenazi, C. Halling, L. Mullin, K. Houmiel, J.Gordon, M. Vaudin, O. Iartchouk, A. Epp, F. Liu, C. Wollam, M. Allinger,D. Doughty, C. Scott, C. Lappas, B. Markelz, C. Flanagan, C. Crowell, J.Gurson, C. Lomo, C. Sear, G. Strub, C. Cielo, and S. Slater. 2001. Genomesequence of the plant pathogen and biotechnology agent Agrobacteriumtumefaciens C58. Science 294:2323–2328.
17. Hacker, S., J. Godeke, A. Lindemann, S. Mesa, G. Pessi, and F. Narberhaus.2008. Global consequences of phosphatidylcholine reduction in Bradyrhizo-bium japonicum. Mol. Genet. Genomics 280:59–72.
18. Hacker, S., C. Sohlenkamp, M. Aktas, O. Geiger, and F. Narberhaus. 2008.Multiple phospholipid N-methyltransferases with distinct substrate specific-ities are encoded in Bradyrhizobium japonicum. J. Bacteriol. 190:571–580.
19. Hanahan, D. 1983. Studies on transformation of Escherichia coli with plas-mids. J. Mol. Biol. 166:557–580.
20. Hindahl, M. S., and B. H. Iglewski. 1984. Isolation and characterization ofthe Legionella pneumophila outer membrane. J. Bacteriol. 159:107–113.
21. Kaneshiro, T., and J. H. Law. 1964. Phosphatidylcholine synthesis inAgrobacterium tumefaciens. I. Purification and properties of a phosphati-dylethanolamine N-methyltransferase. J. Biol. Chem. 239:1705–1713.
22. Kappes, R. M., B. Kempf, S. Kneip, J. Boch, J. Gade, J. Meier-Wagner, andE. Bremer. 1999. Two evolutionarily closely related ABC transporters me-diate the uptake of choline for synthesis of the osmoprotectant glycinebetaine in Bacillus subtilis. Mol. Microbiol. 32:203–216.
23. Karnezis, T., H. C. Fisher, G. M. Neumann, B. A. Stone, and V. A. Stanisich.2002. Cloning and characterization of the phosphatidylserine synthase geneof Agrobacterium sp. strain ATCC 31749 and effect of its inactivation onproduction of high-molecular-mass (133)-�-D-glucan (curdlan). J. Bacte-riol. 184:4114–4123.
24. Klausen, M., A. Heydorn, P. Ragas, L. Lambertsen, A. Aaes-Jørgensen, S.Molin, and T. Tolker-Nielsen. 2003. Biofilm formation by Pseudomonasaeruginosa wild type, flagella and type IV pili mutants. Mol. Microbiol.48:1511–1524.
25. Lambert, A., M. Østerås, K. Mandon, M. C. Poggi, and D. Le Rudulier. 2001.Fructose uptake in Sinorhizobium meliloti is mediated by a high-affinityATP-binding cassette transport system. J. Bacteriol. 183:4709–4717.
26. Liu, A. C., H. W. Shih, T. Hsu, and E. M. Lai. 2008. A citrate-inducible gene,encoding a putative tricarboxylate transporter, is downregulated by the or-ganic solvent DMSO in Agrobacterium tumefaciens. J. Appl. Microbiol. 105:1372–1383.
27. Lopez-Lara, I. M., C. Sohlenkamp, and O. Geiger. 2003. Membrane lipids inplant-associated bacteria: their biosyntheses and possible functions. Mol.Plant-Microbe Interact. 16:567–579.
28. Mandon, K., M. Østerås, E. Boncompagni, J. C. Trinchant, G. Spennato,M. C. Poggi, and D. Le Rudulier. 2003. The Sinorhizobium meliloti glycinebetaine biosynthetic genes (betlCBA) are induced by choline and highlyexpressed in bacteroids. Mol. Plant-Microbe Interact. 16:709–719.
29. Merritt, P. M., T. Danhorn, and C. Fuqua. 2007. Motility and chemotaxis inAgrobacterium tumefaciens surface attachment and biofilm formation. J. Bac-teriol. 189:8005–8014.
30. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring HarborLaboratory Press, Cold Spring Harbor, NY.
31. Minder, A. C., K. E. de Rudder, F. Narberhaus, H. M. Fischer, H. Hennecke,and O. Geiger. 2001. Phosphatidylcholine levels in Bradyrhizobium japoni-cum membranes are critical for an efficient symbiosis with the soybean hostplant. Mol. Microbiol. 39:1186–1198.
32. Osborn, M. J., J. E. Gander, E. Parisi, and J. Carson. 1972. Mechanism ofassembly of the outer membrane of Salmonella typhimurium. Isolation andcharacterization of cytoplasmic and outer membrane. J. Biol. Chem. 247:3962–3972.
33. O’Toole, G. A., and R. Kolter. 1998. Flagellar and twitching motility arenecessary for Pseudomonas aeruginosa biofilm development. Mol. Microbiol.30:295–304.
34. Otto, K., and T. J. Silhavy. 2002. Surface sensing and adhesion of Escherichiacoli controlled by the Cpx-signaling pathway. Proc. Natl. Acad. Sci. USA99:2287–2292.
35. Pratt, L. A., and R. Kolter. 1998. Genetic analysis of Escherichia coli biofilmformation: roles of flagella, motility, chemotaxis and type I pili. Mol. Micro-biol. 30:285–293.
36. Richardson, J. S., M. F. Hynes, and I. J. Oresnik. 2004. A genetic locusnecessary for rhamnose uptake and catabolism in Rhizobium leguminosarumbv. trifolii. J. Bacteriol. 186:8433–8442.
37. Robinson, J. B., O. H. Tuovinen, and W. D. Bauer. 1992. Role of divalentcations in the subunit associations of complex flagella from Rhizobium me-liloti. J. Bacteriol. 174:3896–3902.
38. Sambrook, J., and D. W. Russell. 2001. Molecular cloning: a laboratorymanual, 3rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Har-bor, NY.
VOL. 191, 2009 PC BIOSYNTHESIS IN AGROBACTERIUM TUMEFACIENS 373
at UN
IVE
RS
ITA
T B
OC
HU
M on F
ebruary 19, 2009 jb.asm
.orgD
ownloaded from
33
39. Schafer, A., A. Tauch, W. Jager, J. Kalinowski, G. Thierbach, and A. Puhler.1994. Small mobilizable multi-purpose cloning vectors derived from theEscherichia coli plasmids pK18 and pK19: selection of defined deletions inthe chromosome of Corynebacterium glutamicum. Gene 145:69–73.
40. Schmidt-Eisenlohr, H., N. Domke, C. Angerer, G. Wanner, P. C. Zambryski,and C. Baron. 1999. Vir proteins stabilize VirB5 and mediate its associationwith the T pilus of Agrobacterium tumefaciens. J. Bacteriol. 181:7485–7492.
41. Sherr, S. I., and J. H. Law. 1965. Phosphatidylcholine synthesis in Agrobac-terium tumefaciens. II. Uptake and utilization of choline. J. Biol. Chem.240:3760–3765.
42. Sohlenkamp, C., I. M. Lopez-Lara, and O. Geiger. 2003. Biosynthesis ofphosphatidylcholine in bacteria. Prog. Lipid Res. 42:115–162.
43. Studier, F. W., and B. A. Moffatt. 1986. Use of bacteriophage T7 RNApolymerase to direct selective high-level expression of cloned genes. J. Mol.Biol. 189:113–130.
44. Thormann, K. M., R. M. Saville, S. Shukla, D. A. Pelletier, and A. M.Spormann. 2004. Initial phases of biofilm formation in Shewanella oneidensisMR-1. J. Bacteriol. 186:8096–8104.
45. Wessel, M., S. Klusener, J. Godeke, C. Fritz, S. Hacker, and F. Narberhaus.2006. Virulence of Agrobacterium tumefaciens requires phosphatidylcholinein the bacterial membrane. Mol. Microbiol. 62:906–915.
46. Wilderman, P. J., A. I. Vasil, W. E. Martin, R. C. Murphy, and M. L. Vasil.2002. Pseudomonas aeruginosa synthesizes phosphatidylcholine by use of thephosphatidylcholine synthase pathway. J. Bacteriol. 184:4792–4799.
47. Winans, S. C. 1992. Two-way chemical signaling in Agrobacterium-plantinteractions. Microbiol. Rev. 56:12–31.
48. Wood, D. W., J. C. Setubal, R. Kaul, D. E. Monks, J. P. Kitajima, V. K.Okura, Y. Zhou, L. Chen, G. E. Wood, N. F. Almeida, Jr., L. Woo, Y.Chen, I. T. Paulsen, J. A. Eisen, P. D. Karp, D. Bovee, Sr., P. Chapman,J. Clendenning, G. Deatherage, W. Gillet, C. Grant, T. Kutyavin, R. Levy,M. J. Li, E. McClelland, A. Palmieri, C. Raymond, G. Rouse, C. Saen-phimmachak, Z. Wu, P. Romero, D. Gordon, S. Zhang, H. Yoo, Y. Tao, P.Biddle, M. Jung, W. Krespan, M. Perry, B. Gordon-Kamm, L. Liao, S.Kim, C. Hendrick, Z. Y. Zhao, M. Dolan, F. Chumley, S. V. Tingey, J. F.Tomb, M. P. Gordon, M. V. Olson, and E. W. Nester. 2001. The genomeof the natural genetic engineer Agrobacterium tumefaciens C58. Science294:2317–2323.
49. Yeo, H. J., Q. Yuan, M. R. Beck, C. Baron, and G. Waksman. 2003. Structuraland functional characterization of the VirB5 protein from the type IV se-cretion system encoded by the conjugative plasmid pKM101. Proc. Natl.Acad. Sci. USA 100:15947–15952.
374 KLUSENER ET AL. J. BACTERIOL.
at UN
IVE
RS
ITA
T B
OC
HU
M on F
ebruary 19, 2009 jb.asm
.orgD
ownloaded from
34
Global consequences of the complete loss of PC
35
– C –
Proteomic and transcriptomic characterization of a virulence-deficient phosphatidylcholine-negative
Agrobacterium tumefaciens mutant
Sonja Klüsener, Stephanie Hacker, Yun-Long Tsai, Julia E. Bandow, Ronald Gust,
Erh-Min Lai and Franz Narberhaus
Mol Genet Genomics (2010) 283:575–589
DOI 10.1007/s00438-010-0542-7ORIGINAL PAPER
Proteomic and transcriptomic characterization of a virulence-deWcient phosphatidylcholine-negative Agrobacterium tumefaciens mutant
Sonja Klüsener · Stephanie Hacker · Yun-Long Tsai · Julia E. Bandow · Ronald Gust · Erh-Min Lai · Franz Narberhaus
Received: 24 February 2010 / Accepted: 21 April 2010 / Published online: 1 May 2010© Springer-Verlag 2010
Abstract Phosphatidylcholine (PC) is the most abundantphospholipid in eukaryotic membranes, whereas only a lim-ited number of bacteria are able to synthesize PC. Intrigu-ingly, many of the bacteria with PC-containing membranesinteract with eukaryotic hosts. PC is one of the major mem-brane lipids in the phytopathogenic bacterium Agrobacte-rium tumefaciens. The presence of PC is critical for diversecellular processes like motility, bioWlm formation, stressresistance, and virulence. The exact role of PC in these pro-cesses is unknown. Here, we examined the global conse-quences of the complete loss of PC at the proteomic andtranscriptomic levels. Both strategies validated theimpaired virulence gene induction responsible for the viru-lence defect of the PC-deWcient mutant. In addition, theproteomic approach revealed a limited subset of proteinswith altered abundance including the reduced Xagellar pro-teins FlaA and FlaB, which explains the motility defect of
the PC mutant. At the whole-genome level, the loss of PCwas correlated with altered expression of up to 13% of allgenes, most encoding membrane or membrane-associatedproteins and proteins with functions in the extracytoplasmicstress response. Our integrated analysis revealed thatA. tumefaciens dynamically remodels its membrane proteincomposition in order to sustain normal growth in theabsence of PC.
Keywords Membrane lipids · Phosphatidylcholine · �-Proteobacterium · Agrobacterium · Plant–microbe interaction
Introduction
Phosphatidylcholine (PC, lecithin) is an essential phospho-lipid in eukaryotic cells and plays an important role both asa structural component of membranes and in signal trans-duction processes (Canty and Zeisel 1994). In contrast toeukaryotic cells the most abundant phospholipids in pro-karyotic membranes are phosphatidylglycerol (PG), cardio-lipin (CL), and phosphatidylethanolamine (PE). Based ongenomic sequence analyses it was estimated that onlyapproximately 10% of all bacteria possess putative PC bio-synthesis genes and are, therefore, able to produce PC(Sohlenkamp et al. 2003). The presence of PC was shownexperimentally for ten species (Arondel et al. 1993;Comerci et al. 2006; Conde-Alvarez et al. 2006; Conoveret al. 2008; de Rudder et al. 1997; Hanada et al. 2001;Hindahl and Iglewski 1984; Kaneshiro and Law 1964; Kentet al. 2004; Minder et al. 2001; Sohlenkamp et al. 2000;Tahara et al. 1994; Wessel et al. 2006; Wilderman et al.2002). Many, but not all of these PC-containing bacteriainteract with eukaryotic hosts (Martínez-Morales et al.
Communicated by M. Hecker.
Electronic supplementary material The online version of this article (doi:10.1007/s00438-010-0542-7) contains supplementary material, which is available to authorized users.
S. Klüsener · S. Hacker · J. E. Bandow · F. Narberhaus (&)Ruhr-Universität Bochum, Lehrstuhl für Biologie der Mikroorganismen, Universitätsstrasse 150, NDEF 06/783, 44780 Bochum, Germanye-mail: franz.narberhaus@rub.de
Y.-L. Tsai · E.-M. LaiInstitute of Plant and Microbial Biology, Academia Road, Academia Sinica, Taipei 115, Taiwan
R. GustFreie Universität Berlin, Pharmazeutische Chemie, Königin-Luise-Str. 2+4, 14195 Berlin, Germany
123
36
576 Mol Genet Genomics (2010) 283:575–589
2003; Sohlenkamp et al. 2003). To date, two PC biosynthe-sis pathways are known in bacteria: the methylation path-way and the PC synthase pathway (Sohlenkamp et al.2003). In the methylation pathway, PE is methylated by atleast one phospholipid N-methyltransferase (Pmt) in threesteps to produce PC via the intermediates monometh-ylphosphatidylethanolamine (MMPE) and dimethylphos-phatidylethanolamine (DMPE) using S-adenosylmethione(SAM) as methyl-donor. A similar methylation pathwayexists in eukaryotes, interestingly using evolutionary unre-lated enzymes (Sohlenkamp et al. 2003). The PC synthasepathway is unique to bacteria and forms PC by direct con-densation of choline and CDP-diacylglycerol. This reactionis catalyzed by the membrane-bound enzyme PC synthasePcs (de Rudder et al. 1999).
Both the methylation and the PC synthase pathways con-tribute to PC biosynthesis in the Gram-negative, phytopath-ogenic �-proteobacterium Agrobacterium tumefaciens.Single enzymes for each pathway, PmtA and Pcs, respec-tively, are responsible for PC biosynthesis in this organism(Klüsener et al. 2009; Wessel et al. 2006). The correspond-ing genes are constitutively expressed and are not regulatedby PC, choline availability, or the presence of the virulenceinducer acetosyringone (Klüsener et al. 2009). However,enzyme activity of the A. tumefaciens PmtA is inhibited bythe end products PC and S-adenosylhomocysteine (SAH)(Aktas and Narberhaus 2009).
A PC-negative A. tumefaciens mutant (�pmtA �pcs)shows a dramatic virulence defect, which is based on theimpaired expression of virB and virD genes encoding themembrane-spanning type IV secretion machinery (Wesselet al. 2006) responsible for transfer of the oncogenic T-DNAand eVector proteins into plant cells (Christie and Cascales2005; Winans 1992). Symbiotic plant–microbe interactionsalso strictly depend on the presence of PC in the bacterialpartner. For example, Bradyrhizobium japonicum and Sino-rhizobium meliloti PC biosynthesis mutants are unable toestablish productive nitrogen-Wxing root nodules (de Rudderet al. 2000; Hacker et al. 2008b; Minder et al. 2001;Sohlenkamp et al. 2003). The principal importance of PC forhost–microbe interactions is further underlined by PC-deW-cient mutants of the human pathogens Legionella pneumo-phila and Brucella abortus which are attenuated in virulence(Comerci et al. 2006; Conde-Alvarez et al. 2006; Conoveret al. 2008). Apart from the virulence defect, PC-deWcientA. tumefaciens show other striking phenotypic aberrations.They are hyper-sensitive toward the detergent SDS, show agrowth defect at elevated temperature, and are characterizedby reduced motility and enhanced bioWlm formation(Klüsener et al. 2009; Wessel et al. 2006).
Global approaches revealed multiple changes in the geneexpression proWle in B. japonicum and Bacillus subtiliswhen their membrane composition was altered (Hacker
et al. 2008a; Salzberg and Helmann 2008). To learn moreabout the global eVects of PC-deWciency in A. tumefaciens,we performed proteomic and transcriptomic studies com-paring the wild-type and a PC-deWcient �pmtA �pcsmutant. We expanded the number of aVected virulencegenes and showed that the loss of PC was correlated withaltered expression of multiple genes encoding membrane ormembrane-associated proteins and proteins which fulWlltheir function at the membrane.
Materials and methods
Bacterial strains and growth conditions
All strains and plasmids used in this work are listed inTable 1. Escherichia coli cells were grown at 37°C inLuria–Bertani (LB) medium (Sambrook and Russell 2001)supplemented with 10 �g/ml tetracycline and/or 100 �g/mlampicillin. E. coli DH5� was used as host for all cloningprocedures. A. tumefaciens strain C58 (wild type) andderivative (�pmtA �pcs) were routinely grown at 30°C inYEB complex or AB minimal medium (pH 5.5; 1% (w/v)glucose) (Schmidt-Eisenlohr et al. 1999) supplementedwith 200 �g/ml ampicillin and/or 2 �g/ml tetracycline ifnecessary. To induce virulence gene expression, A. tum-efaciens cells were pre-cultivated in liquid AB minimalmedium to an OD600 nm of 0.2 at 30°C, before addition ofacetosyringone (Sigma–Aldrich, Munich, Germany) to aWnal concentration of 0.1 mM. Cells were further incubatedfor 16 h at 23°C. Control cultures (non-induced) weretreated with solvent (DMSO) only.
Plasmid and mutant construction
Recombinant DNA work was carried out according to stan-dard protocols (Sambrook and Russell 2001). All oligonucle-otides used in this work are listed in Table 2. PCR-generatedfragments of the promoter regions of virB1 (atu6167), tzs(atu6164), modA (atu2561), atu3391, atu3396, atu3692,fecR (atu3693), and atu3675 (Goodner et al. 2001; Woodet al. 2001) were digested with KpnI and XhoI and ligatedinto pAC01 (Liu et al. 2008) treated with the same enzymesto construct transcriptional fusions to the lacZ gene. The cor-rect nucleotide sequence of all constructed plasmids was con-Wrmed by automated sequencing. Plasmids were transferredinto A. tumefaciens via electroporation.
Protein preparation for proteomic analysis
Protein preparation was performed as described previously(Lai et al. 2006) with minor modiWcations. A. tumefaciensC58 wild-type and mutant (�pmtA �pcs) cells were grown
123
37
Mol Genet Genomics (2010) 283:575–589 577
in 300 ml AB minimal medium under non-induced and vir-ulence-induced conditions and harvested by centrifugationat 10,000£g, 4°C, for 10 min. The cells were resuspendedin lysis buVer (50 mM Tris–HCl, pH 7.5; 20% (w/v)sucrose; 0.2 M KCl; 0.2 mM DTT; 0.2 mg/ml DNase I;
0.2 mg/ml RNase A; 1 mM PMSF) to an OD600 nm of 5 anddisrupted by two passes through a chilled French-pressurecell at 16,000 psi. The lysate was treated with 0.5 mg/mllysozyme for 1 h on ice and centrifuged twice at 10,000£g,4°C for 20 min to remove the cell debris. The supernatant
Table 1 Strains and plasmids used in this study
ApR ampicillin, TcR tetracycline
Strain or plasmid Relevant characteristics Reference
Strains
Escherichia coli
DH5� Cloning host Hanahan (1983)
Agrobacterium tumefaciens
C58 Wild type C. Baron, Montreal, Canada
C58 �pmtA �pcs Wild type derivative, deletion of the pmtA and pcs genes Wessel et al. (2006)
Plasmids
pAC01 TcR, ApR; transcriptional lacZ fusion vector containing promoterless lacZ gene
Liu et al. (2008)
pBO392 TcR, ApR; modA-lacZ fusion in pAC01 This study
pBO393 TcR, ApR; atu3391-lacZ fusion in pAC01 This study
pBO395 TcR, ApR; fecR-lacZ fusion in pAC01 This study
pBO396 TcR, ApR; atu3396-lacZ fusion in pAC01 This study
pBO397 TcR, ApR; atu3692-lacZ fusion in pAC01 This study
pBO398 TcR, ApR; atu3675-lacZ fusion in pAC01 This study
pAC01-virBp TcR, ApR; virB1-lacZ fusion in pAC01 This study
pAC01-tzsp TcR, ApR; tzs-lacZ fusion in pAC01 This study
Table 2 Oligonucleotides used in this study
Oligonucleotide Sequence (5� ! 3�)a
Transcriptional lacZ fusions
modA_Fw_KpnI AAAAGGTACCGAGATCGCCGAAG
modA_Rv_XhoI AAAACTCGAGGATGTCCCCCAGG
atu3391_Fw_KpnI AAAAGGTACCGTCTCGGACCAGTC
atu3391_Rv_XhoI AAAACTCGAGCGGATATTGCTCTCC
fecR_Fw_KpnI AAAAGGTACCCGAGGCGCGG
fecR_Rv_XhoI AAAACTCGAGTATGCTCTTCCATGTGTAG
atu3396_Fw_KpnI AAAAGGTACCCCGAAAGGCGAAG
atu3396_Rv_XhoI AAAACTCGAGGTGTTCTCCAGTCG
atu3692_Fw_KpnI AAAAGGTACCCAGACCTTATCGCAGTTC
atu3692_Rv_XhoI AAAACTCGAGCATCCTAATCAACACAGC
atu3675_Fw_KpnI AAAAGGTACCGCTGCCCGCTG
atu3675_Rv_XhoI AAAACTCGAGAAGCCGCCCCTTG
virB_p_KpnI_F AGGGTACCTTGAGGATGTTTTTCAGGGAGACTC
virB_p_XhoI_R GGCTCGAGTGATGTTCACGATTATCGGGAGG
tzs_p_KpnI_F TAGGTACCATCGCCTATCCGCTTT
tzs_p_XhoI_R ATCTCGAGGTGCGCCTCATATAAAT
Control of DNA contamination after RNA isolation
pmtA_Fw_KpnI AAAAGGTACCGATTTTTCTGCTGGCG
pmtA_Rv_XhoI AAAACTCGAGGATTCCCCTCGGATGa Restriction sites in the oligo-nucleotides are italicized
123
38
578 Mol Genet Genomics (2010) 283:575–589
(total protein fraction) was centrifuged at 150,000£g, 4°Cfor 1 h in an ultracentrifuge to fractionate the soluble frac-tion (supernatant) containing cytosolic and periplasmic pro-teins and insoluble fraction (pellet) containing membraneproteins.
To concentrate the soluble protein fraction the resultingsupernatant was TCA (trichloroacetic acid)-precipitated. Tothis end 30 �l of 1% (w/v) DOC (deoxycholate, sodium)was added to 1 ml cytosolic protein fraction, mixed andincubated on ice for 10 min. The proteins were precipitatedwith 100% (w/v) TCA over night at ¡80°C and centrifugedat 16,000£g, 4°C for 10 min. To remove the TCA, the pro-tein pellet was washed with 100% acetone and resuspendedin 1 ml rehydration buVer (7 M urea; 2 M thiourea; 40 mMTris; 1% (w/v) amidosulfobetaine, ASB-14; 1% (v/v) Tri-ton X-100). Protein aggregates were removed by centrifu-gation at 16,000£g for 3 min.
The membrane pellet was carefully washed in 8 ml of50 mM Tris–HCl, pH 7.5 containing 1 mM PMSF bysoniWcation followed by centrifugation at 150,000£g(70.1 Ti), 4°C for 1 h. The membranes were resuspendedin the rehydration buVer described above with 2% (w/v)CHAPS (3-[(3-Cholamidopropyl)dimethylammonio]-1-propanesulfonate) instead of Triton X-100 to eYcientlysolubilize hydrophobic membrane proteins. The samplewas centrifuged at 16,000£g for 3 min to remove insolu-ble membrane components. Protein concentrations in thecytosolic and membrane fractions were determined bythe Bradford assay (Bio-Rad Laboratories GmbH,Munich, Germany).
Two-dimensional (2D) gel electrophoresis
The 2D analysis was performed according to the manufac-turer’s instructions (GE Healthcare Amersham Biosciences,Freiburg, Germany). 100 �g cytosolic proteins and 150 �gmembrane proteins, respectively, were loaded on Immobi-line™ DryStrips (IPG; pH 3–10, nonlinear, 24 cm). Isoelec-tric focusing (IEF) was carried out at steps of 0–500 V(0.5 kVh), 500 V (2.5 kVh), 500–2,000 V (2.5 kVh),2,000 V (2 kVh), 2,000–6,000 V (16 kVh), and 6,000 V(66 kVh) with the Multiphore II system (GE HealthcareAmersham Biosciences). The IPG strips were equilibratedfor 20 min in equilibration buVer A [6 M urea; 2% (w/v)SDS; 30% (v/v) glycerol; 50 mM Tris–HCl, pH 8.8; 1%(w/v) DTT] and 20 min in equilibration buVer B (buVer Awith 3% (w/v) iodoacetamide instead of DTT). Proteinswere separated by their size on 10% SDS-polyacrylamidegels (26 £ 20 cm) at 25°C for 1 h at 0.5 W per gel followedby 4 h at 10 W per gel in an Ettan DALTtwelve system (GEHealthcare Amersham Biosciences). The protein spots werevisualized by RuBP staining as described by Lamanda et al.(2004) with minor modiWcations.
Quantitative image analysis
RuBP-stained gels were scanned on a Typhoon TRIO (GEHealthcare Amersham Biosciences) scanner with the fol-lowing parameters: laser power was set to 550 V, excitationbandwidth of 510–532 nm, and an emission Wlter at610 nm. For the processing of gel images and relative quan-titation of protein spots the software Delta2D Version 4.0(Decodon GmbH, Greifswald, Germany) was used, follow-ing the protocol laid out in (Bandow et al. 2008). Arithme-tic means were calculated based on three biologicalreplicates with two technical replicates each. DiVerentiallyexpressed proteins (>1.5-fold change) were identiWed byMALDI-TOF-MS and MALDI-TOF-TOF Mass Spectrom-etry (Online Resource 1).
In gel digest
The protein spots were excised from stained 2D gels usinga spot cutter (Proteome Works™, Biorad, Hercules, CA,USA) with a picker head of 2 mm diameter. Cut spots weretransferred into 96-well micro titer plates. The tryptic digestwith subsequent spotting on a MALDI-target was carriedout automatically with the Ettan Spot Handling Worksta-tion (GE Healthcare Amersham Biosciences) using the fol-lowing protocol: The gel pieces were washed twice with100 �l of a solution of 50% CH3CN and 50% 50 mMNH4HCO3 for 30 min and once with 100 �l 75% CH3CNfor 10 min. After drying at 37°C for 17 min 10 �l trypsinsolution containing 20 ng/�l trypsin (Promega, Madison,WI, USA) was added and incubated at 37°C for 120 min.For extraction gel pieces were covered with 60 �l 0.1%TFA in 50% CH3CN and incubated for 30 min at 40°C. Thepeptide-containing supernatant was transferred into a newmicro titer plate and the extraction was repeated with 40 �lof the same solution. The supernatants were dried at 40°Cfor 220 min completely. The dry residue was dissolved in2.2 �l of 0.5% TFA in 50% CH3CN, and 0.4 �l of this solu-tion was directly spotted on the MALDI target. Then 0.4 �lof a saturated �-cyano-4-hydroxy cinnamic acid solution in70% CH3CN was added and mixed with the sample byaspirating the mixture Wve times. The samples wereallowed to dry on the target 10–15 min before measurementin MALDI-TOF.
MALDI-TOF mass spectrometry
The MALDI-TOF measurement was carried out on the4800 MALDI TOF/TOF Analyzer (Applied Biosystems,Foster City, CA, USA). This instrument is designed forhigh throughput measurement, being automatically able tomeasure the samples, calibrate the spectra, and analyze thedata using the 4000 Explorer™ Software V3.6.
123
39
Mol Genet Genomics (2010) 283:575–589 579
The spectra were recorded in a mass range from 900 to3,700 Da with a focus mass of 2,000 Da. For one mainspectrum 25 sub-spectra with 100 shots per sub-spectrumwere accumulated using a random search pattern. If theautolytical fragment of trypsin with the mono-isotopic(M + H)+ m/z at 2211.104 reached a signal to noise ratio(S/N) of at least 10, an internal calibration was automati-cally performed as one-point-calibration using this peak.The standard mass deviation was less then 0.15 Da. If theautomatic mode failed (in less then 1%), the calibration wascarried out manually.
After calibration, the peak lists were created by using the“peak to mascot” script of the 4700 Explorer™ Software.Settings were a mass range from 900 to 3,700 Da, a peakdensity of 50 peaks per 200 Da, a minimal area of 100, andmaximal 200 peaks per spot. The peak list was created foran S/N ratio of 6.
MALDI-TOF-TOF mass spectrometry
The MALDI-TOF-TOF measurements were carried out onthe 4800 MALDI TOF/TOF Analyzer (Applied Biosys-tems).
TOF-spectra were acquired for the three strongestpeaks. For one main spectrum 20 sub-spectra with 125shots per sub-spectrum were accumulated using a randomsearch pattern. The internal calibration was automaticallyperformed as one-point-calibration with the mono-isoto-pic Arginine (M + H)+ m/z at 175.119 or Lysine (M + H)+
m/z at 147.107 reached a signal to noise ratio (S/N) of atleast 5.
The peak lists were created by using the script of theGPS Explorer™ Software Version 3.6 (build 329). Settingswere a mass range from 60 to Precursor—20 Da, a peakdensity of 5 peaks per 200 Da, a minimal area of 100 andmaximal 20 peaks per precursor. The peak list was createdfor an S/N ratio of 5.
For data base search the Mascot search engine Version:2.1.04 (Matrix Science Ltd., London, UK) with a speciWcA. tumefaciens sequence database was used.
Western blot analysis
Protein samples were separated by 12.5% SDS-polyacryl-amide gel electrophoresis and transferred to nitrocellulosemembranes (Hybond-C; GE Healthcare Amersham Biosci-ences) using standard protocols (Sambrook and Russell2001). Detection was performed with a chemilumines-cence-based system (Pierce Biotechnology, Rockford,USA) using VirB9 (1:5,000) and Xagellar (1:40,000) pro-tein-speciWc antisera.
RNA puriWcation, cDNA synthesis, labeling, and hybridization
For RNA preparation, A. tumefaciens C58 wild-type andmutant (�pmtA �pcs) cells were grown in 50 ml AB mini-mal medium under non-induced and virulence-inducedconditions for 16 h at 23°C. The cells were rapidly trans-ferred to cold tubes containing 1/10 of the culture volumeof “stop solution” (10% (v/v) Tris–HCl-buVered phenol inethanol, pH 8) and centrifuged at 4,000£g, 4°C for 10 min.The cells were frozen in liquid nitrogen and stored at¡80°C. Total RNA was isolated with the hot phenol-extraction procedure described previously (Babst et al.1996). RNA integrity was checked by agarose gel electro-phoresis. Precipitated RNA (100–200 �g) was treated with100 U of recombinant DNase I (Roche, Mannheim,Germany) for 30 min at 37°C in a reaction volume of110 �l. SUPERase In™ (100 U; Ambion, Huntingdon, UK)was included in the reaction to inhibit potential RNaseactivity. RNA samples were cleaned up with RNeasy spincolumns (Qiagen, Hilden, Germany) and the eluted RNAwas checked again for integrity by agarose gel electropho-resis. The absence of genomic DNA contamination wascontrolled by PCR using the primers pmtA_Fw_KpnI andpmtA_Rv_XhoI (Table 2). cDNA production, cDNA label-ing, hybridization, array scanning, as well as data process-ing and normalization were performed by NimbleGenSystems, Inc. (Madison, USA) according to the Nimble-GenTM Arrays User’s Guide for Gene Expression Analysis(Version 3.0). The 4-plex NimbleGen Gene ExpressionArray for A. tumefaciens (Design ID 080630) represents5,344 genes and is based on NCBI reference sequencesNC_003062, NC_003063, NC_003064 and NC_003065.Each gene is represented by six probes (45mer–60mer oli-gonucleotides), each in two copies on diVerent locations onthe array. Three biological replicates for each cultivationcondition were analyzed for A. tumefaciens C58 wild-typeand �pmtA �pcs mutant strains.
Data processing normalization and further analysis
Data were analyzed using the GeneSpring gene expressionanalysis software Version 7.3 (Agilent). The experimentwas normalized with the default settings for AVymetrixarrays. All values below 0.01 were set to a threshold of 0.01followed by a per-chip normalization to the 50th percentileand a per-gene normalization to median. For statisticalcomparisons, the student t test with a P value threshold of0.05 was applied. Comparing cultivation conditions, geneswere considered to be diVerentially expressed at a foldchange of >§2.
123
40
580 Mol Genet Genomics (2010) 283:575–589
For a functional classiWcation of the diVerentiallyexpressed genes, the JCVI functional gene classiWcation(cellular role: mainrole) for A. tumefaciens (http://cmr.jcvi.org) was used. All entries were categorized accordingto 16 functional categories and an additional categorycontaining genes encoding hypothetical proteins. For sim-plicity we combined classes “unclassiWed”, “unknownfunction”, “mobile and extrachromosomal element func-tions” as well as “signal transduction” within class “othercategories”. Hypergeometric tests (Fisher’s exact test) wereused to identify functional groups that are signiWcantlyover- or under-represented among the diVerentiallyexpressed genes relative to all genes. Only categories witha P value ·0.01 were considered as being signiWcantlyover- or under-represented.
Microarray data accession number
The microarray data have been deposited in the NCBI GeneExpression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo) and are accessible via GEO Series acces-sion number GSE21223.
�-Galactosidase assays
The �-galactosidase activity of A. tumefaciens transcrip-tional fusion strains grown in liquid AB minimal mediumunder non-induced (¡AS) and virulence-induced (0.1 mMAS) conditions was measured according to standard proto-cols (Miller 1972). The plasmid pAC01 (Liu et al. 2008)containing the promoterless lacZ gene was used as negativecontrol and the background activity was below 3 MU.
Graphite furnace atomic absorption spectroscopy (GFAAS)
To measure the intracellular iron concentration of A. tum-efaciens wild type and the �pmtA �pcs mutant, cells weregrown in YEB complex medium or AB minimal mediumsupplemented with 10 �M FeSO4. Cells were washed twicewith and then resuspended in 0.9% (w/v) NaCl to anOD600 nm of 1. The cellular iron content was quantiWedusing GFAAS following previously published experimentalsetups (Gust et al. 1998; Ott et al. 2005, 2007). Cell pelletswere suspended in 1 ml Aqua bidest and lysed by means ofa sonotrode. An aliquot (200 �l) of each lysate wasremoved for protein determination by the Bradford assayand stored at ¡20°C. Samples (200 �l) for iron determina-tion were stabilized by addition of 1% (v/v) Triton X-100and were immediately measured afterwards. Standards forcalibration purposes were prepared as aqueous dilutions ofa commercially available iron standard stock solution(1 mg/ml in Aqua bidest). Results were expressed as �giron per mg cellular protein. A Vario 6 graphite furnace
atomic absorption spectrometer (AnalytikJena AG) wasused for iron quantiWcation. Fe was detected at a wave-length of 248.3 nm with a bandpass of 0.2 nm. A deuteriumlamp was used for background correction. A volume of15 ml thereof was injected into the graphite tubes. Drying,pyrolysis, and atomization in the graphite furnace were per-formed according to already published conditions. Themean AUC (area under curve) absorptions of duplicateinjections were used throughout the study.
Results
Comparison of protein levels in the wild-type and PC-deWcient mutant
We have previously shown that the amount of PC inA. tumefaciens C58 reaches approximately 23% of totalmembrane lipids and that it is located in both, the inner andouter membrane (Klüsener et al. 2009). The global eVectsof the complete loss of PC in the A. tumefaciens �pmtA�pcs mutant were analyzed by comparative 2D gel electro-phoresis (GE). For that purpose, total protein extracts wereisolated from three independent A. tumefaciens C58 wild-type and �pmtA �pcs mutant cultures grown in minimalmedium in the absence or presence of the virulence geneinducer acetosyringone. Fractionated cytosolic and mem-brane proteins were separated by 2D GE followed by stain-ing with RuBP. Representative gels of wild-type and�pmtA �pcs mutant samples under non-induced and viru-lence-induced conditions are presented in Fig. 1 (cytosolicfraction) and Fig. 2 (membrane fraction). From both strains810 protein spots of the cytosolic and 172 protein spots ofthe membrane fraction were quantiWed using Delta2D soft-ware.
In the wild type, seven proteins were signiWcantlyup-regulated and two were down-regulated under viru-lence-induced conditions (Figs. 1 and 2, identiWed proteinsare listed in Tables 3 and 4). In the PC-deWcient mutantonly two proteins were signiWcantly up-regulated and fourwere down-regulated under the same conditions. Comparedwith the wild type, Wve proteins were present at higher lev-els in the PC-deWcient mutant under non-induced and Wveproteins under virulence-induced conditions. In addition,ten proteins were present at lower levels in the PC-deWcientmutant than in the wild type under non-induced and 13 pro-teins under virulence-induced conditions.
Proteins that were diVerentially expressed in all threebiological replicates were excised and subjected toMALDI-MS analysis. Fourty protein spots representing 20gene products were successfully identiWed (Tables 3 and 4).In agreement with previous results (Lai et al. 2006), knownVirA/G-regulated proteins such as VirB8, VirC1, VirE2,
123
41
Mol Genet Genomics (2010) 283:575–589 581
VirH1, and Tzs (Figs. 1, 2; Table 3) were up-regulated inthe wild type under virulence-induced conditions. Consis-tent with our previous Wnding that the type IV secretionsystem is not detectable in the �pmtA �pcs strain (Wesselet al. 2006), VirB8 (one component of the TIVSS) waspresent at lower levels in this mutant (Table 3). Other VirA/G-regulated proteins like VirC1, VirE2, VirH1, and Tzswere not induced by AS in the PC-deWcient mutant(Table 3).
Apart from typical virulence-related proteins, 15 pro-teins with altered abundance in the �pmtA �pcs mutantwere identiWed (Table 4). The most dramatic reduction inthe PC-deWcient mutant was observed for NrdE, the hypo-thetical protein Atu1049, and the Xagella proteins FlaA andFlaB. The latter explains the previously observed non-motile phenotype of the PC-deWcient mutant on minimalagar plates (Klüsener et al. 2009). The reduced Xagella pro-tein levels discovered by proteomic results were conWrmedby Western Blot analysis. Total protein extracts and thecorresponding supernatants from cultures grown under vir-ulence-induced and non-induced conditions were analyzedusing Xagellum (Klüsener et al. 2009) and VirB9 (Baronet al. 2001) speciWc antiserum. The polyclonal Xagellumantiserum detects glycosylated FlaA (»33 kDa), FlaB(33 kDa), and FlaC (32.8 kDa), and the minor componentFlaD (49 kDa) (Deakin et al. 1999; Klüsener et al. 2009).
FlaD was barely detectable in our extracts (data notshown). Under virulence conditions, levels of the other Flaproteins were much higher in the wild type than in the PC-deWcient mutant (Fig. 3a). As a control for successful viru-lence induction, VirB9 was present in wild-type samplesgrown under virulence-induced conditions, whereas it wasabsent in mutant samples (Fig. 3b).
Global transcriptional diVerences between wild-type and PC-deWcient mutant cells
Two-dimensional GE is not the tool of choice to analyzehydrophobic, very acidic, or very basic proteins, nor is it asuitable method to analyze low-abundant proteins (WolVet al. 2008). Thus, a global transcriptomic approach waschosen to complement the proteomic approach. We Wrstanalyzed the overall transcriptomic changes upon viru-lence-induction in the wild type. Of 73 genes with alteredexpression levels 65 genes were up- and eight genes weredown-regulated (Online Resource 2). In order to categorizethe aVected genes, the 5,344 open reading frames repre-sented on the array were classiWed according to the 17 cate-gories issued by the J. Craig Venter Institute (http://cmr.jcvi.org) (Fig. 4).
As expected, when non-induced and virulence-inducedconditions were compared in the wild-type strain, the great
Fig. 1 2D map of cytosolic A. tumefaciens wild-type and �pm-tA �pcs mutant proteins. Pro-teins were isolated from a non-induced and b virulence-induced cultures grown for 16 h at 23°C prior to separation by isoelectric focusing (pH 3–10) followed by 10% SDS-PAGE. Protein spots were visualized by RuBP stain-ing and quantiWed with Delta2D software. DiVerentially expressed proteins are marked with circles and corresponding numbers are listed in Tables 3 and 4
123
42
582 Mol Genet Genomics (2010) 283:575–589
majority of all induced genes (59 out of 73) were encodedon the Ti-plasmid and comprised »30% of all Ti-plasmidgenes represented on the array. The classical vir and viru-
lence-induced genes were found within the most signiW-cantly overrepresented categories “cellular processes” and“viral functions” (Fig. 4a). A comparison of the obtained
Fig. 2 2D map of membrane A. tumefaciens wild-type and �pmtA�pcs mutant proteins. Proteins were isolated from a non-induced andb virulence-induced cultures grown for 16 h at 23°C prior to separationby isoelectric focusing (pH 3–10) followed by 10% SDS-PAGE. Pro-
tein spots were visualized by RuBP staining and quantiWed withDelta2D software. DiVerentially expressed proteins are marked withcircles and corresponding numbers are listed in Tables 3 and 4
Table 3 VirA/G-regulated proteins/genes
MW molecular weight, pI isoelectric point, AS acetosyringone, – indicates the protein/gene was not diVerentially expresseda Protein spots 1–3 are presented in Fig. 1 and protein spots I–II in Fig. 2b Proteins were considered to be diVerentially expressed if the fold change was >1.5 and genes if the fold change was >2 when comparing thediVerent strains under the diVerent cultivation conditions
Noa Name Description MW (kDa) pI Proteome: fold changeb Microarray: fold changeb
+AS vs. –AS �pmtA �pcs vs. WT �pmtA �pcs vs. WT
WT �pmtA �pcs ¡AS +AS ¡AS +AS
1 Tzs Atu6164 Isopentenyl-transferase 27.57 5.6 9.2 – – ¡8.9 ¡8.3 ¡235.0
2 VirE2 Atu6190 ssDNA-binding protein 63.35 6.1 14.0 ¡1.6 – ¡10.5 ¡22.4 ¡215.0
3 VirH1 Atu6150 P450-monoxygenase 47.18 5.7 7.2 – ¡1.6 ¡11.4 ¡8.4 ¡296.0
I VirC1 Atu6180 Coordination of DNA-transfer 25.51 5.0 4.2 ¡1.7 – ¡8.8 ¡5.1 ¡212.0
II VirB8 Atu6174 Type IV secretion system 26.30 7.9 2.3 2.6 ¡2.4 ¡2.1 ¡2.2 ¡211.0
123
43
Mol Genet Genomics (2010) 283:575–589 583
Tab
le4
Vir
ulen
ce-i
ndep
ende
nt p
rote
ins/
gene
s
MW
mol
ecul
ar w
eigh
t; pI
isoe
lect
ric
poin
t; A
S ac
etos
yrin
gone
; – in
dica
tes
the
prot
ein/
gene
was
not
diV
eren
tially
exp
ress
eda
Prot
ein
spot
s 4–
10 a
re p
rese
nted
in F
ig.1
and
pro
tein
spo
ts I
II–X
in F
ig.2
bP
rote
ins
wer
e co
nsid
ered
to
be d
iVer
entia
lly e
xpre
ssed
if
the
fold
cha
nge
was
>1.
5 an
d ge
nes
if t
he f
old
chan
ge w
as >
2 w
hen
com
pari
ng t
he d
iVer
ent
stra
ins
unde
r th
e diV
eren
t cu
ltiva
tion
cond
itio
ns
Noa
Nam
eD
escr
iptio
nM
W (
kDa)
pIP
rote
ome:
fol
d ch
ange
bM
icro
arra
y: f
old
chan
geb
+A
S v
s. ¡
AS
�pm
tA �
pcs
vs. W
T�
pmtA
�pc
s vs
. WT
WT
�pm
tA �
pcs
¡A
S+
AS
¡A
S+
AS
4N
rdE
Atu
0070
Rib
onuc
leos
ide-
diph
osph
ate
redu
ctas
e78
.85
5.9
––
¡6.
5¡
4.4
–2.
6
5A
tu10
49H
ypot
heti
cal p
rote
in25
.26
6.2
¡1.
9¡
1.9
¡7.
4¡
7.1
5.4
5.9
6P
dhA
Atu
1429
Pyr
uvat
e de
hydr
ogen
ase
� su
buni
t33
.62
6.4
––
¡1.
9¡
1.8
4.4
5.2
7D
op A
tu09
77D
O-l
ike
seri
ne p
rote
ase
54.4
45.
3–
–2.
93.
9¡
4.1
–
8A
hcY
Atu
0029
Ade
nosy
lhom
o-cy
stei
nase
51.1
65.
5–
–¡
1.8
––
–
9X
ylF
Atu
3576
AB
C tr
ansp
orte
r (x
ylos
e)36
.06
6.1
––
–¡
1.8
2.5
–
10M
odA
Atu
2561
AB
C tr
ansp
orte
r (m
olyb
date
)27
.77
8.9
––
¡2.
0¡
2.2
3.3
2.6
III
Om
p1 A
tu13
81G
roup
1 ou
ter
mem
bran
e pr
otei
n84
.55
5.1
–¡
2.0
2.0
–4.
13.
0
IVF
adL
Atu
1717
Lon
g-ch
ain
fatty
aci
d tr
ansp
ort p
rote
in45
.29
5.0
¡1.
5–
–2.
49.
64.
1
VN
uoC
Atu
1270
NA
DH
-ubi
quin
one
oxid
ored
ucta
se s
ubun
itC
23.1
14.
7–
–2.
51.
8–
–
VI
Cob
W A
tu45
02C
obal
amin
syn
thes
is p
rote
in42
.04
5.6
––
¡2.
7¡
1.8
––
VII
Nod
T A
tu25
52N
odul
atio
n pr
otei
nT
pre
curs
or51
.24
9.1
1.8
–3.
01.
53.
12.
5
VII
ID
naK
Atu
0122
Cha
pero
ne68
.08
4.9
––
1.6
3.1
––
IXF
laB
Atu
0543
Fla
gella
ass
ocia
ted
prot
ein
32.9
94.
72.
9–
¡3.
5¡
13.7
2.9
3.6
XF
laA
Atu
0545
Fla
gella
ass
ocia
ted
prot
ein
31.6
74.
8–
2.6
¡17
.0¡
8.8
––
123
44
584 Mol Genet Genomics (2010) 283:575–589
gene list (Table 3/Online Resource 2) with previous studies(Anand et al. 2008; Cho and Winans 2005) features a goodcorrelation for classical virulence-induced genes such asthe virA, -B, -C, -D, -E, and tzs operons, the virH, virF, andvirK genes, as well as the repABC operon of the Ti-plas-mid. This underlines the reliability of the NimbleGenA. tumefaciens C58 gene expression array and permits avalid comparison of the PC-deWcient �pmtA �pcs mutantwith the wild type.
In contrast to the wild type, the addition of acetosyrin-gone had no inXuence on the PC-deWcient mutant. Geneexpression was similar under non-induced and virulence-induced conditions (Online Resource 2) conWrming themissing induction of virulence genes observed at the prote-ome level.
Multiple diVerences were observed when the expressionproWles of the PC-deWcient mutant and the wild type werecompared. Approximately 9% (494 genes) and 13% (677genes) of all genes on the array were diVerentiallyexpressed under non-induced and virulence-induced condi-tions, respectively (Online Resource 2). Since virulence-associated genes were not induced, genes belonging to thecategories “cellular processes” and “viral functions” wereoverrepresented among the 318 down-regulated genes(Fig. 4b, c).
While virulence genes were up to 300-fold down-regu-lated in the �pmtA �pcs mutant (Table 3/Online Resource
2), the majority of diVerentially expressed genes unrelatedto virulence were changed only two- to Wve-fold, in rarecases up to 16-fold (Table 4/Online Resource 2). Many ofthose virulence-independent genes are coding for eithermembrane or membrane-associated proteins, evidenced bya statistically signiWcant overrepresentation of genesbelonging to the category “transport and binding proteins”as well as “energy metabolism” (Fig. 4). Among these pro-teins are ABC transport proteins, subunits of the ATP syn-thase, outer membrane proteins, components of type IV piliand newly identiWed type VI secretion systems (Wu et al.2008), as well as two-component sensor kinases (Fig. 5/Online Resource 2). Additionally, we identiWed genes cod-ing for sigma factors of the extracellular function (ECF)family (Fig. 5), proteins involved in exopolysaccharide pro-duction, and cell division (Online Resource 2).
Interestingly, there was no signiWcant overrepresentationof genes classiWed as “phospholipid and fatty acid biosyn-thesis”. Only a few genes from this category showedaltered expression in the �pmtA �pcs strain compared withthe wild type. They encode the fatty acid biosynthesisenzymes FabH (3-oxoacyl-(acyl-carrier-protein) synthaseIII; Atu1179), and FabG (3-oxoacyl-(acyl-carrier protein)reductase; Atu3813) which are both about three-fold up-regulated in the �pmtA �pcs strain. In contrast, no signiW-cant changes were observed for key enzymes of thephospholipid biosynthesis, except for the deleted pmtA andpcs genes themselves. This might indicate that in A. tum-efaciens PC-deWciency and the subsequent changes inmembrane integrity are compensated by changes in thefatty acid composition rather than by changes in thephospholipid headgroup. In a previous study on Saccharo-myces cerevisiae, it has indeed been shown that PC deple-tion is compensated by acyl chain remodeling to maintainthe physical properties of membranes (de Kroon 2007).
An interesting group of genes dramatically induced inthe PC-deWcient mutant encode proteins involved in ironmetabolism, e.g., iron ABC transporters (Atu3388–Atu3391; Atu3394–Atu3396) and heme receptors (Atu2287;Atu3385) (Online Resource 2). As high induction factorsare generally not typical for regulatory genes, it is remark-able that the putative ECF sigma factor gene atu3692 is
Fig. 3 Reduction of Xagellar proteins in the absence of phosphatidyl-choline. A. tumefaciens wild-type and �pmtA �pcs mutant cells weregrown under non-induced (¡AS) and virulence-induced (+AS) condi-tions. a After harvesting, the cell pellet (P) and supernatant (S) wereanalyzed by SDS-PAGE and Western blotting using an antibody(1:40,000) against A. tumefaciens Xagellar proteins (»33 kDa). b Toverify successful virulence induction, pellet samples were analyzed bySDS-PAGE and Western blotting using a VirB9 antibody (1:5,000)against the VirB9 protein (30 kDa)
Fig. 4 DiVerentially expressed genes categorized by functional classi-Wcation according to the JCVI annotation (http://cmr.jcvi.org). Per-centages of up- (black) and down-regulated (grey) genes are given forthe comparison of a wild type virulence-induced versus non-induced,b �pmtA �pcs mutant versus wild type under virulence-induced con-ditions, and c �pmtA �pcs mutant versus wild type under non-inducedconditions. Percentages are calculated by dividing the number of sig-niWcantly up- or down-regulated genes in each category by the totalnumber of genes in the corresponding category. Asterisks indicate sta-tistical signiWcance for overrepresented genes in a given category (Pvalue ·0.01)
�
123
45
586 Mol Genet Genomics (2010) 283:575–589
up-regulated 7- to 16-fold (Fig. 5/Online Resource 2). Like-wise, the downstream-encoded sensor kinase FecR(Atu3693; Fig. 5/Online Resource 2) and the upstream-encoded putative TonB-dependent receptor and ABCtransport proteins (Atu3687–Atu3691) were induced in theabsence of PC. The genetic organization and the predictedprotein functions suggest the presence of a system for ferriccitrate uptake and ferric citrate-dependent gene regulationas it is known from Escherichia coli (Braun 1997).
Independent veriWcation of microarray data by �-galactosidase activity assays
To validate the results from the transcriptomic approach weconstructed plasmid-encoded, transcriptional lacZ-fusionsof selected genes. The resulting plasmids (Table 1) wereelectroporated into A. tumefaciens wild-type and �pmtA�pcs mutant cells. All reporter strains were grown andmonitored under non-induced and virulence-induced condi-tions.
Since the most dramatically aVected genes in theabsence of PC are members of the virulence regulon, weused the virulence genes virB1 and tzs as positive controls.As expected, the virB1-lacZ and tzs-lacZ fusions showedhigh �-galactosidase activity under virulence-induced con-ditions in the wild type, whereas only low levels were mea-sured in the PC-deWcient mutant (Fig. 6). The next setincluded genes whose products are likely located in themembrane such as ModA (ABC transporter, substrate bind-ing protein [SBP]; Atu2561) and/or are involved in ironmetabolism like Atu3391 and Atu3396 (ABC transporter,SBP), FecR (sensor kinase; Atu3693), Atu3692 (ECFsigma factor), and Atu3675 (siderophore biosynthesis).�-Galactosidase activity of modA-lacZ, atu3391-lacZ,
atu3396-lacZ, atu3692-lacZ, and atu3675-lacZ was notdetectable in the wild type independent of growth condi-tions, whereas fecR-lacZ showed signiWcant expressionunder both conditions (Fig. 6). Consistent with thepresented transcriptomic data, all lacZ fusions showedsigniWcant induction of �-galactosidase activity in the PC-deWcient mutant compared with the wild type. The onlyexception was the fecR fusion, which was equallyexpressed under all conditions tested (Fig. 6) while a two-fold induction for fecR was observed in the microarrayexperiment (Fig. 5), suggesting a possible regulation ofmRNA stability.
Fig. 5 Expression of many genes coding for membrane-associated proteins is altered in the �pmtA �pcs mutant. The fold-change in expression of the corresponding genes observed with microarray analysis is giv-en below the respective proteins and an up- or down-regulation is indicated by arrows. Possible signals for sensor kinases are indicated by Xashes. AS acetosy-ringone; Fe iron; IM inner mem-brane; OM outer membrane; P phosphate; RR response regula-tor; SK sensor kinase
Fig. 6 Reporter gene activity of selected genes whose expression wasaltered in the absence of PC in the microarray experiments. A. tumefac-iens wild-type and �pmtA �pcs mutant cells were grown under non-in-duced (¡AS) and virulence-induced (0.1 mM AS) conditions and�-galactosidase activity was determined from various plasmid-encod-ed transcriptional lacZ fusions. The error bars indicate standard devi-ations from three independent assays. 1 wild type; 2 �pmtA �pcs; ECFsigma factor of the extracellular function family; SBP substrate bind-ing protein
123
47
Mol Genet Genomics (2010) 283:575–589 587
Discussion
The global studies presented here provided detailed insightsinto molecular defects elicited by the absence of PC inA. tumefaciens. It emerges that PC mainly aVects bacterialmembrane protein composition. In the absence of PC thecells respond by rearranging their membrane protein pool.The altered composition does not aVect the Wtness ofA. tumefaciens under normal growth conditions but rendersthem more susceptible to stress conditions and, mostimportantly, prevents formation of the type IV secretionsystem and thus impairs plant infection. The overall eVectswere more pronounced as in a B. japonicum pmtA mutantthat still retains a signiWcant amount of PC in the membrane(Hacker et al. 2008a).
Both the proteomic and the transcriptome study con-Wrmed the previously observed impaired induction of VirA/G-controlled virulence genes and the concomitant absenceof the type IV secretion system in the PC mutant (Wesselet al. 2006). Interestingly, some virulence genes showed areduced basal expression in the mutant already under non-induced conditions (68 of 197 Ti-plasmid genes on thearray). This reduction might be explained by the stronglyreduced expression of the central virulence regulator virGin the �pmtA �pcs mutant (Fig. 5, Online Resource 2). Theexpression of virG is regulated via two promoters, P1 andP2. The P1 promoter is stimulated by phenolic compounds(e.g. acetosyringone) as well as phosphate starvation,whereas P2 is controlled by an acid-responsive two-compo-nent system ChvG/I (Charles and Nester 1993; Li et al.2002; Mantis and Winans 1993; Winans 1990; Winanset al. 1988). Presumably, the virG P2 promoter is inducedby low pH resulting in elevated levels of VirG protein. Thesensor kinase VirA, stimulated by acetosyringone, activatesVirG, which then autoregulates transcription of its owngene and all other vir genes. Since neither the expression ofvirA nor that of chvG/I was changed under non-inducedconditions in the PC-deWcient mutant compared with thewild type, the lack of virulence induction is most likely dueto low virG expression in the absence of PC. The nature ofthe virG expression defect is unclear and warrants detailedinvestigation.
Apart from the virulence proteins, the abundance of onlya limited set of other proteins was signiWcantly altered inthe PC mutant. Reduction of the FlaA and FlaB proteins isconsistent with the non-motile phenotype of the PC-deW-cient mutant on minimal medium. A. tumefaciens �XaABCmutants are reported to be non-motile and virulence isattenuated by 38% as compared with the wild type (Ches-nokova et al. 1997). The more severe virulence phenotypeof the �pmtA �pcs mutant already implies that reducedmotility is not the only reason for this defect. The molecu-lar reasons for the motility defect of the PC mutant are not
known to date. Control of Xagellation in A. tumefaciens iscomplex and depends on a number of environmental factors(Lai et al. 2000). We reported previously that Xagellar pro-tein levels are normal when �pmtA �pcs cells grow in com-plex medium (Klüsener et al. 2009). At that time, we didnot examine Xagella from cells grown in minimal mediumbecause we were not aware that the medium compositiondrastically alters Xagellation. Consistent with the presenceof Xagella in complex medium, the PC-deWcient mutant isas motile as the wild type on YEB medium agar plates (datanot shown). One reason for the reduction of Xagella pro-teins in the �pmtA �pcs mutant grown in minimal mediummight be a defective twin-arginine translocation (TAT)pathway. Ding and Christie showed that tatC mutant cellslack detectable Xagella and that the TAT system is impor-tant for Xagellar biosynthesis in A. tumefaciens (Ding andChristie 2003). The reduced level of the periplasmic sub-strate binding protein ModA (Table 4), which is predictedto be a TAT substrate (Ding and Christie 2003), impliesthat the TAT system is not fully active in the �pmtA �pcsmutant under the tested conditions.
A motility defect was also reported for a L. pneumophilaPC-deWcient mutant (Conover et al. 2008). Impaired motil-ity was attributed to a post-transcriptional defect caused bymissing accumulation of Xagella proteins but the molecularbasis for this remains unclear. Since XaA mRNA levelswere not reduced in the A. tumefaciens �pmtA �pcs strain(Table 4), reduction of Fla proteins is likely to be a post-transcriptional event as it is in L. pneumophila.
Notably, many genes involved in iron metabolism werechanged and exhibited induction factors up to 16-fold. Ironis an important trace element in bacteria as it is needed as aco-factor in many enzymes (Schaible and Kaufmann 2004).On the other hand, high intracellular iron concentrationscan generate reactive oxygen species via the Fenton reac-tion resulting in damage of cellular components (Imlayet al. 1988). As a consequence, A. tumefaciens and otherbacteria tightly control intracellular iron levels (Ngok-Ngam et al. 2009), e.g., by the secretion of siderophores,which are low-molecular mass compounds that bind ironwith high aYnity (Schwyn and Neilands 1987). Recently,two operons responsible for the production of siderophoreshave been described in A. tumefaciens C58 (ATS gene clus-ter; atu3670–atu3685) (Rondon et al. 2004). According toour expression studies, the genes for siderophore produc-tion were signiWcantly induced in the PC-deWcient mutant(Fig. 6; atu3675-lacZ).
Since many genes coding for proteins involved in ironmetabolism were induced with high fold changes in theabsence of PC, we measured the intracellular iron concen-tration of the wild type and PC mutant. We detectedmedium-dependent diVerences in cellular iron levels. Theiron content of A. tumefaciens cells was four-fold higher
123
48
588 Mol Genet Genomics (2010) 283:575–589
in minimal medium than in complex medium (data notshown). However, under both conditions, the iron concen-tration was identical in the wild type and the double mutant,suggesting that the net result of the induction of iron uptakesystems in the �pmtA �pcs mutant is a wild-type like ironhomeostasis promoting normal growth.
Acknowledgments We are grateful to Knut Büttner (Greifswald) forMALDI-MS analyses, Hauke Hennecke and co-workers (Zürich) forproviding access to the GeneSpring gene expression analysis software,Birgit Scharf (Blacksburg, Virginia) for antiXagella sera, and ChristianBaron (Montreal) for VirB9 antisera. We thank Yi-Chun Chen (Taipei)for constructing pAC01-virBp and pAC01-tzsp, Meriyem Aktas andSina Langklotz for helpful comments on this manuscript, and Chris-tiane Fritz for technical assistance. The work was in part supported bya grant the German Research Foundation (DFG NA 240/7) to FN, agrant from Taiwan National Science Council (NSC 95-2320-B-001-009) to EML and a joint grant from the German Academic ExchangeService (DAAD) and the Taiwan National Science Council (PPP grantno. 0970029248P) to FN and EML.
References
Aktas M, Narberhaus F (2009) In vitro characterization of the enzymeproperties of the phospholipid N-methyltransferase PmtA fromAgrobacterium tumefaciens. J Bacteriol 191:2033–2041
Anand A, Uppalapati SR, Ryu CM, Allen SN, Kang L, Tang Y,Mysore KS (2008) Salicylic acid and systemic acquired resistanceplay a role in attenuating crown gall disease caused by Agrobac-terium tumefaciens. Plant Physiol 146:703–715
Arondel V, Benning C, Somerville CR (1993) Isolation and functionalexpression in Escherichia coli of a gene encoding phosphatidyl-ethanolamine methyltransferase (EC 2.1.1.17) from Rhodobactersphaeroides. J Biol Chem 268:16002–16008
Babst M, Hennecke H, Fischer HM (1996) Two diVerent mechanismsare involved in the heat-shock regulation of chaperonin geneexpression in Bradyrhizobium japonicum. Mol Microbiol19:827–839
Bandow JE, Baker JD, Berth M, Painter C, Sepulveda OJ, Clark KA,Kilty I, VanBogelen RA (2008) Improved image analysis workXowfor 2-D gels enables large-scale 2-D gel-based proteomics stud-ies—COPD biomarker discovery study. Proteomics 8:3030–3041
Baron C, Domke N, Beinhofer M, Hapfelmeier S (2001) Elevatedtemperature diVerentially aVects virulence, VirB protein accumu-lation, and T-pilus formation in diVerent Agrobacteriumtumefaciens and Agrobacterium vitis strains. J Bacteriol183:6852–6861
Braun V (1997) Surface signaling: novel transcription initiation mech-anism starting from the cell surface. Arch Microbiol 167:325–331
Canty DJ, Zeisel SH (1994) Lecithin and choline in human health anddisease. Nutr Rev 52:327–339
Charles TC, Nester EW (1993) A chromosomally encoded two-com-ponent sensory transduction system is required for virulence ofAgrobacterium tumefaciens. J Bacteriol 175:6614–6625
Chesnokova O, Coutinho JB, Khan IH, Mikhail MS, Kado CI (1997)Characterization of Xagella genes of Agrobacterium tumefaciens,and the eVect of a bald strain on virulence. Mol Microbiol23:579–590
Cho H, Winans SC (2005) VirA and VirG activate the Ti plasmidrepABC operon, elevating plasmid copy number in response towound-released chemical signals. Proc Natl Acad Sci USA102:14843–14848
Christie PJ, Cascales E (2005) Structural and dynamic properties ofbacterial type IV secretion systems (review). Mol Membr Biol22:51–61
Comerci DJ, Altabe S, de Mendoza D, Ugalde RA (2006) Brucellaabortus synthesizes phosphatidylcholine from choline providedby the host. J Bacteriol 188:1929–1934
Conde-Alvarez R, Grilló MJ, Salcedo SP, de Miguel MJ, Fugier E,Gorvel JP, Moriyón I, Iriarte M (2006) Synthesis of phosphatidyl-choline, a typical eukaryotic phospholipid, is necessary for fullvirulence of the intracellular bacterial parasite Brucella abortus.Cell Microbiol 8:1322–1335
Conover GM, Martínez-Morales F, Heidtman MI, Luo ZQ, Tang M,Chen C, Geiger O, Isberg RR (2008) Phosphatidylcholine synthe-sis is required for optimal function of Legionella pneumophilavirulence determinants. Cell Microbiol 10:514–528
de Kroon AI (2007) Metabolism of phosphatidylcholine and its impli-cations for lipid acyl chain composition in Saccharomyces cere-visiae. Biochim Biophys Acta 1771:343–352
de Rudder KE, Thomas-Oates JE, Geiger O (1997) Rhizobium melilotimutants deWcient in phospholipid N-methyltransferase still con-tain phosphatidylcholine. J Bacteriol 179:6921–6928
de Rudder KE, Sohlenkamp C, Geiger O (1999) Plant-exuded cholineis used for rhizobial membrane lipid biosynthesis by phosphati-dylcholine synthase. J Biol Chem 274:20011–20016
de Rudder KE, López-Lara IM, Geiger O (2000) Inactivation of thegene for phospholipid N-methyltransferase in Sinorhizobiummeliloti: phosphatidylcholine is required for normal growth.Mol Microbiol 37:763–772
Deakin WJ, Parker VE, Wright EL, Ashcroft KJ, Loake GJ, Shaw CH(1999) Agrobacterium tumefaciens possesses a fourth Xagelingene located in a large gene cluster concerned with Xagellar struc-ture, assembly and motility. Microbiology 145:1397–1407
Ding Z, Christie PJ (2003) Agrobacterium tumefaciens twin-arginine-dependent translocation is important for virulence, Xagellation,and chemotaxis but not type IV secretion. J Bacteriol 185:760–771
Goodner B, Hinkle G, Gattung S et al (2001) Genome sequence of theplant pathogen and biotechnology agent Agrobacterium tumefac-iens C58. Science 294:2323–2328
Gust R, Schnurr B, Krauser R, Bernhardt G, Koch M, Schmid B,Hummel E, Schönenberger H (1998) Stability and cellular studiesof [rac-1,2-bis(4-Xuorophenyl)-ethylenediamine][cyclobutane-1,1-dicarboxylato]platinum(II), a novel, highly active carboplatinderivative. J Cancer Res Clin Oncol 124:585–597
Hacker S, Gödeke J, Lindemann A, Mesa S, Pessi G, Narberhaus F(2008a) Global consequences of phosphatidylcholine reduc-tion in Bradyrhizobium japonicum. Mol Genet Genomics280:59–72
Hacker S, Sohlenkamp C, Aktas M, Geiger O, Narberhaus F (2008b)Multiple phospholipid N-methyltransferases with distinct sub-strate speciWcities are encoded in Bradyrhizobium japonicum.J Bacteriol 190:571–580
Hanada T, Kashima Y, Kosugi A, Koizumi Y, Yanagida F, Udaka S(2001) A gene encoding phosphatidylethanolamine N-methyl-transferase from Acetobacter aceti and some properties of its dis-ruptant. Biosci Biotechnol Biochem 65:2741–2748
Hanahan D (1983) Studies on transformation of Escherichia coli withplasmids. J Mol Biol 166:557–580
Hindahl MS, Iglewski BH (1984) Isolation and characterization of theLegionella pneumophila outer membrane. J Bacteriol 159:107–113
Imlay JA, Chin SM, Linn S (1988) Toxic DNA damage by hydrogenperoxide through the Fenton reaction in vivo and in vitro. Science240:640–642
Kaneshiro T, Law JH (1964) Phosphatidylcholine synthesis inAgrobacterium tumefaciens. I. PuriWcation and properties of a
123
49
Mol Genet Genomics (2010) 283:575–589 589
phosphatidylethanolamine N-methyltransferase. J Biol Chem239:1705–1713
Kent C, Gee P, Lee SY, Bian X, Fenno JC (2004) A CDP-choline path-way for phosphatidylcholine biosynthesis in Treponema dentico-la. Mol Microbiol 51:471–481
Klüsener S, Aktas M, Thormann KM, Wessel M, Narberhaus F (2009)Expression and physiological relevance of Agrobacterium tum-efaciens phosphatidylcholine biosynthesis genes. J Bacteriol191:365–374
Lai EM, Chesnokova O, Banta LM, Kado CI (2000) Genetic and envi-ronmental factors aVecting T-pilin export and T-pilus biogenesisin relation to Xagellation of Agrobacterium tumefaciens. J Bacte-riol 182:3705–3716
Lai EM, Shih HW, Wen SR, Cheng MW, Hwang HH, Chiu SH (2006)Proteomic analysis of Agrobacterium tumefaciens response to thevir gene inducer acetosyringone. Proteomics 6:4130–4136
Lamanda A, Zahn A, Roder D, Langen H (2004) Improved RutheniumII tris (bathophenantroline disulfonate) staining and destainingprotocol for a better signal-to-background ratio and improvedbaseline resolution. Proteomics 4:599–608
Li L, Jia Y, Hou Q, Charles TC, Nester EW, Pan SQ (2002) A globalpH sensor: Agrobacterium sensor protein ChvG regulates acid-inducible genes on its two chromosomes and Ti plasmid. ProcNatl Acad Sci USA 99:12369–12374
Liu AC, Shih HW, Hsu T, Lai EM (2008) A citrate-inducible gene,encoding a putative tricarboxylate transporter, is downregulatedby the organic solvent DMSO in Agrobacterium tumefaciens.J Appl Microbiol 105:1372–1383
Mantis NJ, Winans SC (1993) The chromosomal response regulatorygene chvI of Agrobacterium tumefaciens complements an Esche-richia coli phoB mutation and is required for virulence. J Bacte-riol 175:6626–6636
Martínez-Morales F, Schobert M, López-Lara IM, Geiger O (2003)Pathways for phosphatidylcholine biosynthesis in bacteria.Microbiology 149:3461–3471
Miller JH (1972) Experiments in molecular genetics. Cold Spring Har-bour Laboratory Press, Cold Spring Harbour
Minder AC, de Rudder KE, Narberhaus F, Fischer HM, Hennecke H,Geiger O (2001) Phosphatidylcholine levels in Bradyrhizobiumjaponicum membranes are critical for an eYcient symbiosis withthe soybean host plant. Mol Microbiol 39:1186–1198
Ngok-Ngam P, Ruangkiattikul N, Mahavihakanont A, Virgem SS,Sukchawalit R, Mongkolsuk S (2009) Roles of Agrobacteriumtumefaciens RirA in iron regulation, oxidative stress response,and virulence. J Bacteriol 191:2083–2090
Ott I, Schmidt K, Kircher B, Schumacher P, Wiglenda T, Gust R(2005) Antitumor-active cobalt-alkyne complexes derived fromacetylsalicylic acid: studies on the mode of drug action. J MedChem 48:622–629
Ott I, SchäZer H, Gust R (2007) Development of a method for thequantiWcation of the molar gold concentration in tumour cellsexposed to gold containing drugs. ChemMedChem 2:702–707
Rondon MR, Ballering KS, Thomas MG (2004) IdentiWcation andanalysis of a siderophore biosynthetic gene cluster from Agrobac-terium tumefaciens C58. Microbiology 150:3857–3866
Salzberg LI, Helmann JD (2008) Phenotypic and transcriptomic char-acterization of Bacillus subtilis mutants with grossly alteredmembrane composition. J Bacteriol 190:7797–7807
Sambrook J, Russell DW (2001) Molecular cloning: a laboratory manual,3rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor
Schaible UE, Kaufmann SH (2004) Iron and microbial infection. NatRev Microbiol 2:946–953
Schmidt-Eisenlohr H, Domke N, Angerer C, Wanner G, ZambryskiPC, Baron C (1999) Vir proteins stabilize VirB5 and mediate itsassociation with the T pilus of Agrobacterium tumefaciens. J Bac-teriol 181:7485–7492
Schwyn B, Neilands JB (1987) Universal chemical assay for the detec-tion and determination of siderophores. Anal Biochem 160:47–56
Sohlenkamp C, de Rudder KE, Röhrs VV, López-Lara IM, Geiger O(2000) Cloning and characterization of the gene for phosphatidyl-choline synthase. J Biol Chem 275:18919–18925
Sohlenkamp C, López-Lara IM, Geiger O (2003) Biosynthesis ofphosphatidylcholine in bacteria. Prog Lipid Res 42:115–162
Tahara Y, Yamashita T, Sogabe A, Ogawa Y (1994) Isolation andcharacterization of Zymomonas mobilis mutant defective in phos-phatidylethanolamine N-methyltransferase. J Gen Appl Micro-biol 40:389–396
Wessel M, Klüsener S, Gödeke J, Fritz C, Hacker S, Narberhaus F(2006) Virulence of Agrobacterium tumefaciens requires phos-phatidylcholine in the bacterial membrane. Mol Microbiol62:906–915
Wilderman PJ, Vasil AI, Martin WE, Murphy RC, Vasil ML (2002)Pseudomonas aeruginosa synthesizes phosphatidylcholine by useof the phosphatidylcholine synthase pathway. J Bacteriol184:4792–4799
Winans SC (1990) Transcriptional induction of an Agrobacterium reg-ulatory gene at tandem promoters by plant-released phenoliccompounds, phosphate starvation, and acidic growth media.J Bacteriol 172:2433–2438
Winans SC (1992) Two-way chemical signaling in Agrobacterium–plant interactions. Microbiol Rev 56:12–31
Winans SC, Kerstetter RA, Nester EW (1988) Transcriptional regula-tion of the virA and virG genes of Agrobacterium tumefaciens.J Bacteriol 170:4047–4054
WolV S, Hahne H, Hecker M, Becher D (2008) Complementary anal-ysis of the vegetative membrane proteome of the human pathogenStaphylococcus aureus. Mol Cell Proteomics 7:1460–1468
Wood DW, Setubal JC, Kaul R et al (2001) The genome of the naturalgenetic engineer Agrobacterium tumefaciens C58. Science294:2317–2323
Wu HY, Chung PC, Shih HW, Wen SR, Lai EM (2008) Secretomeanalysis uncovers an Hcp-family protein secreted via a type VIsecretion system in Agrobacterium tumefaciens. J Bacteriol190:2841–2850
123
50
VirB-induced small heat-shock protein HspL
51
– D –
Small heat-shock protein HspL is induced by VirB protein(s) and promotes VirB/D4-mediated DNA transfer in
Agrobacterium tumefaciens
Yun-Long Tsai, Ming-Hsuan Wang, Chan Gao, Sonja Klüsener, Christian Baron,
Franz Narberhaus and Erh-Min Lai
Small heat-shock protein HspL is induced by VirBprotein(s) and promotes VirB/D4-mediated DNAtransfer in Agrobacterium tumefaciens
Yun-Long Tsai,1 Ming-Hsuan Wang,1 Chan Gao,3 Sonja Klusener,2
Christian Baron,3,4 Franz Narberhaus2 and Erh-Min Lai1
Correspondence
Erh-Min Lai
emlai@gate.sinica.edu.tw
1Institute of Plant and Microbial Biology, Academia Sinica, Taipei, Taiwan
2Lehrstuhl fur Biologie der Mikroorganismen, Ruhr-Universitat Bochum, Bochum, Germany
3Biology Department, McMaster University, Hamilton, ON, Canada
4Departement de Biochimie, Universite de Montreal, Montreal, QC, Canada
Received 7 May 2009
Revised 10 June 2009
Accepted 21 June 2009
Agrobacterium tumefaciens is a Gram-negative plant-pathogenic bacterium that causes crown
gall disease by transferring and integrating its transferred DNA (T-DNA) into the host genome. We
characterized the chromosomally encoded alpha-crystallin-type small heat-shock protein (a-Hsp)
HspL, which was induced by the virulence (vir) gene inducer acetosyringone (AS). The
transcription of hspL but not three other a-Hsp genes (hspC, hspAT1, hspAT2) was upregulated
by AS. Further expression analysis in various vir mutants suggested that AS-induced hspL
transcription is not directly activated by the VirG response regulator but rather depends on the
expression of VirG-activated virB genes encoding components of the type IV secretion system
(T4SS). Among the 11 virB genes encoded by the virB operon, HspL protein levels were reduced
in strains with deletions of virB6, virB8 or virB11. VirB protein accumulation but not virB
transcription levels were reduced in an hspL deletion mutant early after AS induction, implying that
HspL may affect the stability of individual VirB proteins or of the T4S complex directly or indirectly.
Tumorigenesis efficiency and the VirB/D4-mediated conjugal transfer of an IncQ plasmid
RSF1010 derivative between A. tumefaciens strains were reduced in the absence of HspL. In
conclusion, increased HspL abundance is triggered in response to certain VirB protein(s) and
plays a role in optimal VirB protein accumulation, VirB/D4-mediated DNA transfer and
tumorigenesis.
INTRODUCTION
Agrobacterium tumefaciens is a soil-borne plant-pathogenicbacterium causing crown gall disease in a wide range ofplants through an interkingdom DNA delivery system. A.tumefaciens is capable of sensing plant-released woundsignal molecules such as sugars and phenolic compoundsto activate a signal transduction pathway for infection. Thisevent is regulated by the VirA/VirG two-component systemencoded by the tumour-inducing (Ti) plasmid in conjunc-tion with ChvE, a chromosomally encoded periplasmicgalactose/glucose-binding protein to activate the expres-sion of virulence (vir) genes and operons, including virA, B,G, C, D and E (McCullen & Binns, 2006). The transferred
DNA (T-DNA) is then processed, followed by transfer ofthe T-complex and effector proteins via the Ti plasmid-encoded Vir type IV secretion system (T4SS) from bacteriainto the host plant cells (Baron, 2006; Christie et al., 2005).
The Ti Vir T4SS is a transmembrane complex consisting ofVirD4 and 11 VirB proteins that also assembles T pili(Christie, 2001; Lai & Kado, 2000). Accumulating bio-chemical and genetic data suggest a model of an orderedVirB/D4 T4SS assembly pathway (Baron, 2006; Christieet al., 2005; Ward et al., 2002). First, VirB8 initiates T4SSassembly and targets VirB1 to the cell pole, where it maylocally lyse the cell wall to facilitate T4SS assembly acrossthe double membranes (Judd et al., 2005; Yuan et al.,2005). VirB6, VirB4, VirB7, VirB8, VirB9 and VirB10 thenassemble a core complex, which is followed by recruitmentof the subunits important for pilus assembly, includingVirB2, VirB3 and VirB5 (Krall et al., 2002). The VirB11homohexameric ATPase may supply energy for VirB2polymerization across the periplasm to form the T pilus
Abbreviations: AS, acetosyringone; a-Hsp, a-crystallin-type small heat-shock protein; RFU, relative fluorescence units; T4SS, type IV secretionsystem; T-DNA, transferred DNA; Ti, tumour-inducing (plasmid).
A supplementary table of primers and a supplementary figure showingthe reduced tumorigenesis efficiency of the hspL mutant are availablewith the online version of this paper.
Microbiology (2009), 155, 3270–3280 DOI 10.1099/mic.0.030676-0
52
(Atmakuri et al., 2004; Rashkova et al., 2000). Finally,VirB4 and VirD4 are required for substrate translocation,which may be mechanistically linked to a conformationalchange of VirB10 (Cascales & Christie, 2004a). The T-DNA/substrate is translocated via four discrete steps ofsequential interactions with VirD4, VirB11, VirB6/VirB8and VirB2/VirB9, as demonstrated by T-DNA immuno-precipitation assay (Cascales & Christie, 2004b).Biochemical approaches have identified subassemblies ofVirB proteins constituting the ‘core’ components believedto form the translocation channel and the ‘pilus assembly’complex comprising pilus components and associatedfactors (Krall et al., 2002; Yuan et al., 2005). In additionto transporting the T-complex and effector proteins frombacteria into plant cells, the VirB/D4 T4SS can translocatean incompatibility group Q (IncQ) plasmid RSF1010 fromA. tumefaciens into plant cells (Buchanan-Wollaston et al.,1987) or between agrobacteria (Beijersbergen et al., 1992).Hitherto, little has been known about the contribution(s)of non-VirB proteins to the function of the T4SS; the workpresented here suggests a role for the small heat-shockprotein HspL.
We have previously used proteomics approaches to identifyacetosyringone (AS)-induced proteins and discovered ASinduction of HspL (Lai et al., 2006). HspL is an alpha-crystallin-type small heat-shock protein (a-Hsp) thatcontains a characteristic a-crystallin domain (Narberhaus,2002). a-Hsps are a diverse protein family of low-molecular-mass chaperones that exist universally in mostorganisms, including animals, plants, bacteria and archaea.Rhizobiaceae contain a large number of a-Hsp genes, butlittle is known about their function except for the heatshock induction and chaperone-like activities of some ofthem; that is, they prevent model substrates from heat-induced aggregation (Munchbach et al., 1999a, b; Rosenet al., 2002; Studer & Narberhaus, 2000). In A. tumefaciens,there are at least four a-Hsp genes: hspC (atu0375) encodedon the circular chromosome, hspL (atu3887) encoded onthe linear chromosome and hspAT1 (atu5052) and hspAT2(atu5449), both encoded on the pAT plasmid (Balsigeret al., 2004). The latter three are induced by heat shock,and heat induction of hspL is regulated by rpoH, whichencodes an alternative s32-like transcription factor (Rosenet al., 2002; Balsiger et al., 2004)
In this study, we characterized the regulation of HspL andits function in the virulence of A. tumefaciens. The resultsindicate that AS-induced HspL protein accumulation isregulated in a VirB-dependent manner. Further molecularand functional analyses suggest that HspL protein isrequired for optimal VirB protein accumulation, whichmay be important for efficient VirB/D4-mediated DNAtransfer and virulence.
METHODS
Bacterial strains and growth conditions. Bacterial strains andplasmids used in this study are listed in Table 1. For vir gene
induction, A. tumefaciens cells grown overnight at 25 uC in 523 broth(Kado & Heskett, 1970) with appropriate antibiotics were harvestedby centrifugation (8000 g, 10 min) and resuspended in fresh I-medium (AB-MES, pH 5.5) (Lai & Kado, 1998) without antibiotics,at OD600 ~0.1. After growth at 25 uC to OD600 ~0.2, the cells werefurther cultured at 25 uC for different times in the presence of200 mM acetosyringone (AS) (Sigma-Aldrich) (0.1 %, v/v, of 200 mMstock dissolved in DMSO) until harvesting. The controls were grownin the same conditions without any treatment or treated with DMSO,the solvent used to dissolve AS. The concentrations of antibiotics usedwere: 100 mg ampicillin (Ap) ml21, 20 mg tetracycline (Tc) ml21 and10 mg gentamicin (Gm) ml21 for Escherichia coli; 50 mg erythromycin(Em) ml21, 50 mg rifampicin (Rm) ml21, 250 mg spectinomycin (Sp)ml21, 20 mg Tc ml21 and 50 mg Gm ml21 for A. tumefaciens.
Plasmid construction. Details of the primers used in this study aregiven in Supplementary Table S1, available with the online version ofthis paper. The techniques used for DNA cloning and PCR followedstandard protocols (Sambrook & Russell, 2001). Plasmid DNA wasisolated using the Plasmid Miniprep Purification kit provided byGeneMark. To construct plasmids for promoter activity assay, DNAfragments of the hspL, hspC, hspAT1, hspAT2 and virB promoterregions were amplified by PCR and digested with SpeI and HindIIIprior to ligation into the promoter-probe vector pRU1156 at the samesites. The translational fusion between the first three amino acids ofHspL and GFP (HspLD4–160-GFP) was constructed by ligatingHindIII/SpeI-digested hspLt PCR product and SpeI/XbaI-digestedgfp PCR product into the HindIII/XbaI site of pRU1156. To generatethe hspL deletion mutant, the plasmids pEML651 and pEML776 wereconstructed for gene replacement experiments. Plasmid pEML651 wasconstructed by ligating PstI/EcoRI-digested HspL1 PCR product(upstream of hspL), EcoRI-digested GmR gene cassette and EcoRI/SalI-digested HspL2 PCR product (downstream of hspL), into thePstI/SalI sites of the suicide vector pEML649. Plasmid pEML649 wasgenerated by ligating a BamHI-digested sacB PCR product intopJM22 at the BamHI site. Plasmid pEML776 was constructed byligating PstI/EcoRI-digested HspL1 PCR product and EcoRI/SalI-digested HspL2 PCR product into the PstI/SalI sites of the suicidevector pJQ200KS. Plasmid pHspL was constructed by ligating aHindIII-digested HspL PCR product (containing promoter and ORF)into the same site of pEML652 for complementation test. To produceHis-tagged HspL proteins, the DNA fragment containing the hspLORF without the stop codon was amplified by PCR with specificprimers, digested with NdeI and XhoI, and inserted at the same siteof pET-22b(+) to result in plasmid pETHspL. The plasmidconstructs obtained were confirmed by restriction mapping andDNA sequencing.
GFP quantification. To quantify the GFP activities of A. tumefacienscells expressing a gfp transcriptional or translational fusion, thebacterial cells were collected and normalized to OD600 0.2 with 0.9 %NaCl. A 100 ml cell suspension was loaded into a Nunc F96MicroWell plate and analysed for GFP fluorescence with a multilabelplate reader (Chameleon; Hidex Ltd) at 535/485 nm for emission andexcitation. Promoter activity was expressed as relative fluorescenceunits (RFU) after subtracting the fluorescence signal detected from avector control strain and normalized at OD600 0.1.
Real-time RT-PCR. A. tumefaciens strain NT1RE(pJK270) wasgrown for 16 h at 25 uC in I-medium with the addition of DMSOor AS. Total RNA was extracted (Zuber & Losick, 1983) and subjectedto reverse transcription with SuperScript III RNase H2 ReverseTranscriptase (Invitrogen) (Lai et al., 2006) with the appropriate 39
primers (Supplementary Table S1). All primers were designed withthe software PRIMER EXPRESS 2.0 (Applied Biosystems). PCR wasperformed in 25 ml SYBR Master Mix with 100 ng template cDNAand use of an ABI PRISM 7900 HT Sequence Detection System
VirB-induced small heat-shock protein HspL
53
(Applied Biosystems) according to the methods described previously
(Lin et al., 2008). To compare data from different PCR runs or cDNA
samples, CT values for all target genes were normalized to the CT value
of 16S rRNA, a constitutively expressed gene with approximately
equal PCR efficiency in cells treated with AS or DMSO.
Gene replacement by double crossover. Plasmid pEML651 wasused to generate the hspL deletion mutant with replacement of theGmR gene cassette, and plasmid pEML776 was used to generate themarkerless hspL deletion mutant in A. tumefaciens strain NT1RE.Plasmid pE1962-Sp was used to generate NT1RE-Sp, the recipient
Table 1. Bacterial strains and plasmids
Strain/plasmid Relevant characteristics Reference/source
A. tumefaciens
A136 RmR, strain C58 cured of the pTiC58 plasmid Watson et al. (1975)
A348 RmR, A136 containing octopine-type Ti plasmid pTiA6 Garfinkel et al. (1981)
PC1001–PC1011 RmR, A348 derivatives each containing a virB gene deletion from pTiA6 Berger & Christie (1994)
NT1RE RmR EmR, C58 cured of its pTiC58 Watson et al. (1975)
NT1RE-Sp RmR EmR SpR, NT1RE containing spectinomycin-resistant gene (aadA) This study
NT1RE(pJK270) RmR EmR KmR/NmR; pJK270 is pTiC58TraC with Tn5 insertion in the T-DNA region
without affecting virulence
Kao et al. (1982)
NT1RE(pJK107) RmR EmR KmR/NmR, virA : : Tn5 in pTiC58TraC, virG mutant Kao et al. (1982)
NT1RE(pJK105) RmR EmR KmR/NmR, virD1 : : Tn5 in pTiC58TraC, virD1 polar mutant Rogowsky et al. (1987)
NT1RE(pJK502) RmR EmR KmR/NmR, virB3 : : Tn5 in pTiC58TraC, virB3 polar mutant Lundquist et al. (1984)
NT1RE(pJK505) RmR EmR KmR/NmR, virE : : Tn5 in pTiC58TraC, virE polar mutant Rogowsky et al. (1987)
NT1RE(pJK702) RmR EmR KmR/NmR, virC1 : : nptII in pTiC58TraC, virC1 mutant Rogowsky et al. (1987)
NT1RE(pJK710) RmR EmR KmR/NmR, virG : : nptII in pTiC58TraC, virG mutant Kao et al. (1982)
NT1RE(pEL1000) RmR EmR KmR/NmR, virB operon deletion in pJK270 This study
EML770 RmR EmR KmR/NmR GmR, hspL replaced by Gm cassette to generate hspL deletion
mutant in NT1RE(pJK270)
This study
EML815 RmR EmR KmR/NmR GmR TcR, pHspL in EML770 for complementation experiment This study
EML1057 RmR EmR KmR/NmR, markerless hspL deletion mutant in NT1RE(pJK270) This study
EML1280 RmR EmR KmR/NmR TcR, pHspL in EML1057 for complementation experiment This study
E. coli
DH10B Host for DNA cloning Invitrogen
S-17 Host for conjugation Simon et al. (1983)
Plasmids
pGEMT-Easy ApR, TA cloning vector Promega
pJQ200KS GmR, plasmid containing GmR and sacB gene for selection of double crossover Quandt & Hynes (1993)
pUCGMV1 GmR, broad-host-range GmR cassettes for site-specific insertion Schweizer (1993)
pGEMT-Sp ApR SpR, pGEM-T-easy vector containing spectinomycin-resistance gene (aadA) cassette Wu et al. (2008)
pET-22b(+) ApR, an E. coli overexpression vector to generate C-terminal His-tagged protein Novagen
pRU1064 ApR TcR, stable broad-host-range promoter-probe vector containing gfpUV and gusA Karunakaran et al. (2005)
pRU1156 ApR TcR, stable broad-host-range promoter-probe vector containing gfpmut3.1 and gusA Karunakaran et al. (2005)
pRUhspLp ApR TcR, expression of PhspL-gfp transcriptional fusion on pRU1156 This study
pRUhspCp ApR TcR, expression of PhspC-gfp transcriptional fusion on pRU1156 This study
pRUhspAT1p ApR TcR, expression of PhspAT1-gfp transcriptional fusion on pRU1156 This study
pRUhspAT2p ApR TcR, expression of PhspAT2-gfp transcriptional fusion on pRU1156 This study
pRUvirBp ApR TcR, expression of PvirB-gfp transcriptional fusion on pRU1156 This study
pRUhspLt ApR TcR, expression of PhspL-gfp translational fusion protein HspLD4–160-GFP on
pRU1156
This study
pJM22 KmR, vector for M2(Flag) epitope tagging Janine Maddock
pEML649 KmR, pJM22 containing sacB gene This study
pEML651 KmR GmR, up- and downstream fragments of hspL and GmV cassette inserted at
PstI/SalI site of pEML649
This study
pEML652 ApR TcR, pRU1064 digested with PstI to remove reporter gene (gfpUV and gusA) Liu et al. (2008)
pEML776 GmR, up- and downstream fragments of hspL inserted at PstI/SalI site of pJQ200KS This study
pHspL ApR TcR, expression of hspL gene containing its promoter and ORF on pEML652 This study
pETHspL ApR, overexpression of HspL-His in E. coli This study
pML122DKm IncQ plasmid RSF1010 derivative pML122 with removal of nptII gene Labes et al. (1990)
pE1962 TcR, plasmid for introducing gene into the pgl/picA locus in the chromosome of
A. tumefaciens
Lee et al. (2001)
pE1962-Sp TcR SpR, pE1962 containing SpR gene cassette This study
Y.-L. Tsai and others
54
strain for RSF1010 conjugation assay, by transfer of the SpR genecassette into the pgl/picA locus of A. tumefaciens strain NT1RE. A 5 mlvolume of overnight culture (grown in LB broth without antibiotics)of E. coli strain S-17 containing the respective plasmid and A.tumefaciens strain NT1RE were mixed and incubated at 28 uC on LBagar overnight. The bacterial cells were then streaked out on LB agarcontaining Em, Rm and Km and incubated at 28 uC for 2 days toobtain the first crossover events. Three colonies were randomlyselected and streaked out on the same selection medium for furthercolony purification. Each of three independent colonies was grown in5 ml I-medium without antibiotics at 20 uC overnight; serialdilutions (up to 1022) were plated onto 523 agar containing 5 %(w/v) sucrose and incubated at 20 uC for 4–7 days. The colonies werethen selected for the respective antibiotic resistance and confirmed bycolony PCR. The Ti plasmid pJK270 was transferred into theconfirmed mutants by conjugation.
HspL antibody production. The overexpression of HspL-His followedthe instructions of the pET user manual (Novagen, EMD Biosciences).HspL-His was purified with use of Ni-NTA His Bind resins (Novagen),following the manufacturer’s instructions. A 1 mg sample of purifiedHspL-His protein was separated by 15 % (w/v) glycine SDS-PAGE(Sambrook & Russell, 2001), followed by Coomassie brilliant blue R-250staining (Sambrook & Russell, 2001). The major 19 kDa protein,corresponding to the putative HspL-His, was cut out for polyclonalantibody production in rabbits (GlycoNex, Taipei, Taiwan).
Western blot analysis. Proteins were resolved by glycine SDS-PAGE(Sambrook & Russell, 2001) or Tricine SDS-PAGE (Schagger & vonJagow, 1987). Western blot analysis was performed as describedpreviously (Lai & Kado, 1998) with use of primary polyclonalantibodies against HspL, VirB (Baron et al., 2001; Shirasu & Kado,1993), VirD4 (Chen & Kado, 1996), VirE2 (Baron et al., 2001) andneomycin phosphotransferase II (NptII) (Sigma-Aldrich) followed bysecondary antibody using horseradish peroxidase (HRP)-conjugatedgoat anti-rabbit IgG (chemichem) and detection by the use of theWestern Lightning System (Perkin Elmer). Chemiluminescent signalswere visualized on X-ray film (Kodak).
Tumour assay on potato tuber discs. Quantitative tumorigenesisassays with potato tuber discs were as described previously(Shurvinton & Ream, 1991; Wu et al., 2008) except that bacterialcells at OD600 0.4–0.6 were collected and resuspended in PBS at 108
and 107 c.f.u. ml21 for inoculation. The potato tuber discs wereplaced on water agar, infected with 10 ml bacterial culture andincubated at 24 uC for 2 days. Discs were then placed on water agarsupplemented with 100 mg timentin ml21 to kill bacteria andincubated at 24 uC for 3 weeks before tumours were scored.
Conjugal transfer analysis of IncQ plasmid RSF1010. Theconjugation assay was as described by Fullner & Nester (1996) withminor modifications. The donor strains were NT1RE(pJK270) and itsderivatives and the recipient strain was NT1RE-Sp. Cultures of donorand recipient strains were grown overnight at 25 uC in 523 broth withantibiotics. The cells of donor and recipient strains were collected bycentrifugation (8000 g, 5 min) and resuspended in fresh I-mediumwithout antibiotics to OD600 ~0.1. After growth at 25 uC with shakingfor 6 h, 200 mM AS was added to the cultures, which continued togrow at 25 uC for an additional 2 h. Donor and recipient cells weremixed together at a ratio of 10 : 1, and 10 ml of the mating mix wasspotted on sterilized 1 cm2 nylon paper placed on I-medium agar inthe presence of 200 mM AS. After incubation at 25 uC for 3 days, thecells from the nylon paper were resuspended in 1 ml 0.9 % NaCl. Thebacterial suspensions with or without dilution were plated onto 523agar supplemented with appropriate antibiotics and incubated at28 uC for 2 days to select the transconjugants (EmR, GmR, SpR), inputdonors (EmR, GmR), and recipients (EmR, SpR). The number of
transconjugants present in the selected input donors was ignoredbecause of their low frequency (~1025) in the population.
RESULTS
AS-induced hspL expression
The small heat-shock protein HspL was previously identifiedas an AS-induced protein in A. tumefaciens (Lai et al., 2006).To determine whether AS also induces the three other a-Hspgenes (hspC, hspAT1, hspAT2), we analysed the promoteractivity of each a-Hsp gene transcriptionally fused to gfp.The PhspL-gfp transcriptional fusion was upregulated 1.5- to2-fold in cells grown in the presence of AS for 16 h or 24 h ascompared with the non-induced (DMSO) controls (Fig. 1a).In contrast, the promoter activities of hspC, hspAT1 andhspAT2 were not affected by AS treatment. Quantitative RT-PCR, analysis of translational fusion and Western blotanalysis were carried out to monitor the individual steps ofhspL expression. We detected about 5- to 6-fold increasedhspL mRNA level (Fig. 1b), 2-fold increase of HspLD4–160-GFP translational efficiency (Fig. 1c), and 50-fold higherHspL protein level (Fig. 2a) in cells grown in the presence ofAS for 16 h in comparison to the DMSO or H2O controls.These data suggest that AS-induced transcription of hspL is aspecific response rather than a general effect of AS treatmenton heat-shock gene expression.
HspL protein accumulation is induced by AS in aVirB protein-dependent manner
Similar to other known Vir proteins that are regulated bythe VirA/VirG two-component system, AS-induced HspLaccumulation is also dependent on virA and virG (Lai et al.,2006). Interestingly, in contrast to the presence of VirG-binding sequences (vir box) in the promoters of known virregulon genes (Cho & Winans, 2005), the apparent absenceof a vir box within the putative hspL promoter regionsuggests that hspL is not directly activated by the VirGresponse regulator. To address this question, the PhspL-gfptranscriptional fusion construct was transformed intodifferent vir mutants. As expected, hspL promoter activitywas upregulated by AS in the wild-type and the virC, virDand virE mutants, but not in the virA or virG mutants, aftereither 16 h or 40 h AS induction (Fig. 2b). Surprisingly,hspL promoter activity was also compromised in a polarvirB3 mutant (Fig. 2b) and the steady-state level of AS-induced hspL mRNA was also diminished in the virB3mutant (data not shown). Furthermore, HspL proteinlevels were increased by AS in the wild-type and the virC,virD and virE mutants but not in the polar virB3 mutant(Fig. 2a). The VirA/VirG two-component system is stillfunctional in the virB3 polar mutant because other Virproteins such as VirD4 and VirE2 were still induced by AS(Fig. 2a). Taken together, these data suggest that AS-induced hspL transcription is not directly activated by theresponse regulator VirG but rather is linked to theexpression of virB genes.
VirB-induced small heat-shock protein HspL
55
To determine which virB gene(s) are responsible for AS-induced HspL protein accumulation, individual virBnonpolar deletion mutants were analysed, and the HspLprotein was detected by Western blot analysis. Reduced levelsof HspL protein were observed in the virB1 and virB2mutants, but only at 16 h after AS induction (Fig. 3). Moreclearly, after 16 h and 40 h of AS induction, HspL proteinlevels were reduced in virB6, virB8 and virB11 deletion strainsas compared with the wild-type and the other virB mutants.VirE2 levels, as a control, were normal in these strains. Thesedata suggest that HspL protein accumulation is likelyinduced by the expression of one or a subset of VirB proteins.
The absence of HspL causes reduced VirB proteinaccumulation at an early stage of AS inductionwithout affecting virB transcription
We investigated the physiological role of HspL proteinexpression and accumulation in response to certain VirBproteins. Small heat-shock proteins are generally thoughtto bind selectively to non-native proteins to prevent theiraggregation and degradation (Narberhaus, 2002; Sun &
MacRae, 2005). Thus, HspL might play a role as achaperone in stabilizing VirB proteins and therebycontribute to efficient T-DNA transfer and tumorigenesis.To test this hypothesis, we determined the effect of deletionof hspL on VirB protein accumulation. The loss of HspL inthe deletion mutant and its complementation by expres-sion of hspL (driven by its native promoter) on an IncPplasmid was demonstrated by Western blotting (Fig. 4a).The level of all VirB proteins analysed was lower in thehspL deletion mutant than in the wild-type soon after ASinduction (4 h and 8 h). In contrast, levels of other AS-induced proteins such as VirD4 and VirE2 and the internalcontrol protein NptII were not substantially reduced,which suggests that AS-induced HspL acts mainly on VirBproteins. The effects were complemented by hspL expres-sion in trans (Fig. 4a). In contrast, hspL plays no role in AS-induced virB transcription because the virB promoter wasinduced by AS at similar levels in both the wild-type andthe hspL mutant up to 16 h (Fig. 4b). Taken together, thedata suggest that HspL plays a role in optimal VirB proteinaccumulation likely via maintaining their stability duringthe assembly process.
700
600
500
400
300
200
100
GFP
sig
nal (
RFU
)
16 h 40 hDMSOAS
DMSOAS
P hspL P hspC P hspAT1 P hspAT2 P hspL P hspC P hspAT1 P hspAT2
6
5
4
3
2
1
Rel
ativ
e qu
antit
y of
hsp
L m
RN
A
H2ODMSOAS
H2ODMSOAS
1000
800
600
400
200
GFP
sig
nal (
RFU
)
(a)
(b) (c)
Fig. 1. AS-induced hspL expression. (a) Relative GFP signal of A. tumefaciens strain NT1RE(pJK270) containing a gfp
transcriptional fusion to the promoter of hspL, hspC, hspAT1 or hspAT2. The bacteria were grown at 25 6C for 16 or 40 h inI-medium with the addition of DMSO or AS. (b) Quantitative RT-PCR analysis of the hspL mRNA level of strain NT1RE(pJK270)and (c) relative GFP signal of strain NT1RE(pJK270) expressing HspLD4–160-GFP fusion protein (driven by its native promoter),grown at 25 6C for 16 h in I-medium without (H2O) or with DMSO or AS. The relative GFP signals for promoter activity areshown as the mean±SD of three independent experiments.
Y.-L. Tsai and others
56
The absence of HspL causes reduced VirB/D4-mediated DNA transfer and tumorigenesisefficiency
In view of the reduced VirB protein accumulation in thehspL deletion mutant, we were curious to determinewhether HspL is involved in the VirB/D4-mediated DNAtransfer and virulence of A. tumefaciens. An IncQ plasmid,
RSF1010, can be transferred between strains of A.tumefaciens by the Ti plasmid-encoded VirB/D4 machineryin an AS-induced environment (Beijersbergen et al., 1992).We used this alternative functional T4SS assay todetermine whether hspL contributes to the conjugaltransfer using the RSF1010-derived plasmidpML122DKm. The transfer efficiency was consistentlyreduced, by 30 % on average, in the hspL mutant as
1000
800
600
400
200GFP
sig
nal (
RFU
)
16 h 40 h
(a)
(b)DMSOAS
WT virA virB virC virD virE virG WT virA virB virC virD virE virG
Fig. 2. AS-induced HspL protein accumula-tion is regulated in a VirB-dependent manner.(a) Western blot analysis of HspL, VirB9,VirD4, VirE2 and NptII when wild-type anddifferent vir mutants were grown at 25 6C for16 or 40 h in I-medium with addition of DMSOor AS. (b) Relative GFP signals of A. tumefa-
ciens strains containing plasmid pRUhspLp,expressing the PhspL-gfp transcriptionalfusion, grown at 25 6C for 16 or 40 h inI-medium with DMSO or AS. Strains: WT,NT1RE(pJK270); virA, NT1RE(pJK107); virB,NT1RE(pJK502); virC, NT1RE(pJK702); virD,NT1RE(pJK105); virE, NT1RE(pJK505); virG,NT1RE(pJK710). The relative GFP signals forpromoter activity are shown as the mean±SD
of three independent experiments. Westernblotting was performed for at least threeindependent experiments with similar results.NptII protein levels were determined as thecontrols.
Fig. 3. HspL protein accumulation is compro-mised in the virB6, virB8 and virB11 nonpolardeletion mutants: Western blot analysis of HspLand VirE2 in A. tumefaciens strains containingoctopine-type Ti plasmid pTiA6 and its variants.A348, the wild-type strain, was treated with AS(+) or DMSO (”) and each of the virB non-polar deletion mutants (DB1–DB11 representdeletions of virB1 to virB11) induced by ASwere determined. Western blotting was per-formed for at least three independent experi-ments with similar results. VirE2 protein levelswere determined as the controls.
VirB-induced small heat-shock protein HspL
57
compared with the wild-type (Table 2). The mobilizationability of the hspL deletion mutant was restored and evenincreased by complementation with an hspL-expressingplasmid, which indicates that HspL contributes to efficientmobilization of pML122DKm between A. tumefaciensstrains. The observed mobilization of pML122DKm wasindeed mediated by the Ti VirB/D4 T4SS because notransconjugants were detected in the mutant with deletionof the entire virB operon (Table 2) or when the conjugationexperiment was carried out in the absence of AS (data notshown). We also determined whether the deletion of hspLaffects the function of another Ti plasmid-encoded T4SS,the trb locus, which mediates the conjugal transfer of the Tiplasmid between agrobacteria (Li et al., 1998; von Bodmanet al., 1989). The conjugal transfer efficiency of pTiC58-derived pJK270 was not affected in the hspL deletionmutant or in its complemented strain as compared with inthe wild-type (data not shown), which suggests that HspLis not a general factor involved in DNA and plasmid
transfer. Thus, HspL promotes RSF1010 conjugal transfer,which further substantiates its role in Ti plasmid-encodedVirB/D4 T4SS-dependent function.
To determine the effects on A. tumefaciens virulence,quantitative tumorigenesis assays on potato tuber discswere used to determine the effect of HspL on tumourformation. The tumorigenesis efficiency was consistently20–25 % lower in the hspL mutant than in the wild-typewith inoculation of 107 and 108 c.f.u. ml21 of bacterial cells(Fig. 5). Complementation of hspL from a plasmid in thehspL deletion mutant restored tumour formation to wild-type levels, which indicated that the attenuated virulence ofthe hspL mutant was caused specifically by the loss of hspL(Fig. 5). The reduced tumorigenesis efficiency caused bythe deletion of hspL was also observed in infectedArabidopsis thaliana roots (Supplementary Fig. S1). Thesedata strongly suggest that HspL contributes to promoteVirB/D4-mediated DNA transfer and disease development.
34kDa
23
16
7
23
16
Wild-type (AS)Wild-type (DMSO)DhspL (AS)DhspL (DMSO)
800
600
400
200
0
GFP
sig
nal (
RFU
)
40 8 12 16Time (h)
(a)
(b)
Fig. 4. The absence of HspL causes reducedVirB protein accumulation at early stages of ASinduction without affecting virB operon tran-scription. (a) Wild-type NT1RE(pJK270),DhspL (hspL deletion mutant, EML770) andDhspL(pHspL) (complemented strain,EML815) were grown in I-medium at 25 6Cin the presence of AS to induce vir geneexpression. The total cell lysates were sub-jected to Tricine-SDS-PAGE followed byWestern blot analysis. Numbers on the rightare molecular masses of reference proteins inkDa. NptII served as an internal control. Atleast three independent experiments werecarried out with similar results. (b) RelativeGFP signal of A. tumefaciens NT1RE(pJK270)or DhspL (hspL deletion mutant, EML770)containing plasmid pRUvirBp expressing thePvirB-gfp transcriptional fusion. The bacteriawere grown at 25 6C in I-medium with theaddition of DMSO or AS and collected atdifferent times for GFP quantification. Therelative GFP signals for hspL promoter activityare shown as the mean±SD of at least threeindependent experiments.
Y.-L. Tsai and others
58
DISCUSSION
Most of the components required for conjugal DNAtransfer and tumour formation by A. tumefaciens areencoded on the Ti plasmid. In this study, we demonstratedthat the chromosomally encoded small heat-shock proteinHspL is induced upon expression of certain VirB proteins,the major components of the Ti VirB/D4 T4SS required forgene transfer to plants in A. tumefaciens. Our genetic andfunctional evidence suggests a role for HspL in promoting
VirB protein stability, VirB/D4-mediated DNA transferand tumorigenesis.
The VirB-induced HspL expression resembles the induc-tion of heat-shock proteins and proteases via the extra-cytoplasmic (or envelope) stress responses observed inmany Gram-negative bacteria (Raivio, 2005; Rowley et al.,2006). The known envelope stress response is regulated viathe CpxAR two-component regulatory system or thealternative sigma factor sE (Raivio, 2005). In E. coli, theexpression and assembly of functional plasmid R1-determined T4SS pili and of type IV bundle-forming piliwere found to elicit the envelope stress responses via theCpxAR two-component regulatory system (Nevesinjac &Raivio, 2005; Zahrl et al., 2006). The discovery of VirB-induced HspL in this study and the identification of Cpxregulation of a-Hsp genes ibpA and ibpB in E. coli (Lau-Wong et al., 2008) suggest the involvement of a-Hsps inenvelope stress responses. To our knowledge, however,HspL is the first a-Hsp demonstrated to be involved in thefunction of a protein secretion system. The mechanism ofhspL induction is not clear because CpxAR componentscould not be identified in the A. tumefaciens C58 genomebased on BLAST analysis. RpoH (s32) is required for heat-shock-induced HspL protein accumulation (Rosen et al.,2002) and likely regulated at the transcriptional level due tothe presence of an RpoH-dependent promoter of hspL(Balsiger et al., 2004). However, RpoH was not essential forAS-induced HspL protein accumulation under non-heat-shock conditions because the HspL protein level accumu-lated to the wild-type level in the A. tumefaciens rpoHmutant after AS induction at 25 uC (data not shown).Therefore, the expression of certain VirB proteins maytrigger an as yet unknown regulator(s) that upregulate(s)hspL expression in A. tumefaciens under non-heat-shockconditions.
Although the assembly of pili by the IncFII plasmid R1T4SS triggered the envelope stress response and a decreasedresponse was observed in a traA pilin mutant (Zahrl et al.,2006), the exact T4SS component that mediates itsinduction is unknown. Our data indicate that AS-inducedHspL protein accumulates to the wild-type level in most ofthe single virB deletion mutants after 40 h induction(Fig. 3), which suggests that T4SS-induced HspL proteinaccumulation requires neither the presence of an intact
Table 2. Effect of hspL on mobilization of pML122DKm in A. tumefaciens
Donor strain Relevant genotype Conjugation frequency (%)*
Experiment 1 Experiment 2 Experiment 3
NT1RE(pEL1000) virB operon deletion mutant ,4.3561028 (,0.27) ,3.6161028 (,0.29) ,4.1461028(,0.21)
NT1RE(pJK270) Wild-type 1.6361025 (100) 1.2661025 (100) 1.9661025 (100)
EML1057 DhspL 1.0861025 (66.02) 7.9661026 (63.2) 1.5661025 (79.3)
EML1280 DhspL(pHspL) 2.9861025 (182.25) 1.6461025 (130.2) 4.0161025 (204.0)
*The conjugation frequency is expressed as number of transconjugants per input donor.
Wild-type DhspL DhspL(pHspL)
107
107 108
108
30
25
20
15
10
5No.
of t
umou
rs p
er d
isc
25.4
±1.
4
18.9
±0.
8
25.7
±1.
5
23.4
±1.
4
18.6
±0.
9
22.7
±1.
2
Wild-typeDhspL DhspL(pHspL)
(a)
(b)
Fig. 5. Quantitative tumorigenesis assay on potato tuber discs. (a)Wild-type NT1RE(pJK270), DhspL (hspL deletion mutant,EML770) and DhspL(pHspL) (complemented strain, EML815)were examined for their tumorigenesis efficiency on potato tuberdiscs by inoculation at 108 and 107 c.f.u. ml”1. Tumorigenesisefficiency is expressed as the number of tumours per disc(mean±SE, calculated from results of 60 potato tuber discs foreach analysed strain in each independent experiment). (b)Representative results; at least four independent experimentswere carried out with similar results.
VirB-induced small heat-shock protein HspL
59
secretion system nor the formation of the T pilus. Theevidence that the HspL protein level was markedly reducedin the virB6, virB8 and virB11 deletion mutants suggeststhat HspL protein accumulation may be triggered by oneor a subset of T4SS components. We noticed that VirB8and VirB11 protein levels were reduced in the virB6deletion mutant (data not shown). Since deletion of virB6had a negative effect on downstream gene expression(virB7–virB11) (Liu & Binns, 2003), it remains to bedetermined whether the requirement of VirB6 in triggeringHspL protein accumulation is direct or indirect. VirB6,VirB8 and VirB11 are inner-membrane componentsdirectly involved in the T-DNA/substrate translocationpathway (Cascales & Christie, 2004b); however, T-DNAtranslocation through this T4SS channel is not required forHspL induction because the VirD4 coupling protein isdispensable for this effect (Fig. 2a, b). Interestingly, thedeletion of virB8 caused the greatest decrease in HspLprotein level (Fig. 3). Because VirB8 is an assembly factorthat may initiate T4SS assembly (Baron, 2006), wespeculate that hspL transcription and its protein accu-mulation may be triggered by the formation of the earlysubassembly complex.
Although the VirB protein level was clearly reduced, we didnot observe effects on virB transcription in the hspLdeletion mutant as compared with the wild-type (Fig. 4),which suggests that HspL may function as a VirBchaperone. Interestingly, although hspL seems to beexpressed at a basal level and is upregulated only abouttwofold by AS at the transcriptional level, based on ourpromoter activity assay (Fig. 1a) and microarray data(Anand et al., 2008), HspL protein is barely detectable inthe absence of AS but accumulates markedly – about 50-fold – upon AS induction, in a VirB protein-dependentmanner (Fig. 2a). The twofold upregulation of theHspLD4–160-GFP translational fusion by AS (Fig. 1c) furthersuggests a post-translational regulation of AS-inducedHspL accumulation. Both a chaperone and its interactingsubstrate become stabilized when they interact with eachother (Narberhaus, 2002; Sun & MacRae, 2005). Thus, wespeculate that HspL protein might be stabilized wheninteracting with its substrates such as VirB proteins andmay be rapidly degraded in the absence of its substrate.Likewise, the substrates may be more susceptible toproteolysis in the absence of the chaperone. We arecurrently investigating whether HspL interacts directly withVirB protein(s) and functions as a VirB chaperone toprevent VirB from aggregation and degradation, therebymaintaining the stability and/or functionality of theindividual VirB proteins and/or the assembled T4SScomplexes.
The discovery of HspL as a non-VirB factor contributing toT4SS protein stability is novel, but most importantly, thedecreased VirB protein accumulation in the absence ofHspL also correlates with the reduced tumorigenesisefficiency of the hspL mutant as compared with the wild-type (Fig. 5, Supplementary Fig. S1). Obviously HspL plays
a specific role for the Ti VirB/D4 T4SS because VirB/D4-mediated RSF1010 transfer but not Trb-mediated Tiplasmid transfer between agrobacteria was decreased inthe absence of HspL (Table 2). This specificity was furthersupported by evidence that hspL but none of the otherthree a-Hsp genes (hspC, hspAT1 and hspAT2) wasupregulated by AS (Fig. 1a) and no deleterious effects ongrowth or membrane lipid composition were detected inthe absence of hspL (data not shown). However, one mayargue that HspL does not contribute an essential functionfor A. tumefaciens virulence because the reduced VirBprotein accumulation and tumorigenesis efficiency was notas drastic in the hspL mutant as the wild-type (Fig. 4a;compare with Fig. 5, Supplementary Fig. S1). Functionalredundancy of a-Hsps was found in E. coli, in which thesimultaneous presence of a-Hsps IbpA and IbpB enhancedthe stabilization of thermally aggregated proteins ascompared with the presence of IbpA or IbpB alone(Matuszewska et al., 2005). Thus, it is possible that thebasal-level expression of the other three a-Hsps maypartially substitute for the function of HspL in VirBprotein stability and T4SS function in the absence of HspL.Examining the effect on the stability of VirB proteins/complexes, VirB/D4-mediated DNA transfer and tumor-igenesis in single or multiple a-Hsp mutants would be aninteresting future study.
In general, the expression of bacterial a-Hsp genes is low orundetectable under normal growth conditions but isinduced to high levels under heat shock or other stressconditions (Narberhaus, 2002). The induction of a-Hspgenes was previously reported during bacterial infectionwith the human pathogens Mycobacterium tuberculosis andMycobacterium leprae. The a-Hsp genes acr1 and acr2 areinduced in M. tuberculosis-infected macrophages and acr2was further demonstrated to be required for the patho-genicity (Stewart et al., 2005, 2006; Wilkinson et al., 2005).The expression of another a-Hsp gene, shsp18, encoding asurface-exposed antigen of M. leprae (Ilangumaran et al.,1994; Lini et al., 2008), was activated in macrophages(Dellagostin et al., 1995). Our findings that the phyto-pathogen A. tumefaciens exploits the VirB-induced HspLexpression to promote its tumorigenesis add to the list ofa-Hsp participation in bacterial virulence. Further studycould explore the importance of small heat-shock proteinsin the virulence of other bacterial pathogens and elucidatethe molecular mechanisms underlying their regulation andinvolvement in their infection processes under non-heat-shock conditions.
ACKNOWLEDGEMENTS
We are grateful to Clarence Kado for providing several A. tumefaciens
strains and antibodies of VirB2, VirB9 and VirD4. We also thankPeter Christie for providing A348-derived virB mutants.
Acknowledgements also go to Shu-Hsing Wu, Kuo-Chen Yeh andthe lab members for discussion and technical assistance. This work
was supported by the Taiwan National Science Council (NSC) grantNSC 95-2320-B-001-009 to E.-M. L. and an NSC postdoctoral
Y.-L. Tsai and others
60
fellowship to Y.-L. T, by grant MOP-84239 from the Canadian
Institutes of Health Research (CIHR) to C. B., and a DFG grant(German Research Foundation, SFB 480) to F. N.
REFERENCES
Anand, A., Uppalapati, S. R., Ryu, C. M., Allen, S. N., Kang, L., Tang, Y.& Mysore, K. S. (2008). Salicylic acid and systemic acquired resistance
play a role in attenuating crown gall disease caused by Agrobacteriumtumefaciens. Plant Physiol 146, 703–715.
Atmakuri, K., Cascales, E. & Christie, P. J. (2004). Energeticcomponents VirD4, VirB11 and VirB4 mediate early DNA transfer
reactions required for bacterial type IV secretion. Mol Microbiol 54,
1199–1211.
Balsiger, S., Ragaz, C., Baron, C. & Narberhaus, F. (2004). Replicon-
specific regulation of small heat shock genes in Agrobacterium
tumefaciens. J Bacteriol 186, 6824–6829.
Baron, C. (2006). VirB8: a conserved type IV secretion system
assembly factor and drug target. Biochem Cell Biol 84, 890–899.
Baron, C., Domke, N., Beinhofer, M. & Hapfelmeier, S. (2001).Elevated temperature differentially affects virulence, VirB protein
accumulation, and T-pilus formation in different Agrobacteriumtumefaciens and Agrobacterium vitis strains. J Bacteriol 183, 6852–6861.
Beijersbergen, A., Dulk-Ras, A. D., Schilperoort, R. A. & Hooykaas,P. J. (1992). Conjugative transfer by the virulence system ofAgrobacterium tumefaciens. Science 256, 1324–1327.
Berger, B. R. & Christie, P. J. (1994). Genetic complementationanalysis of the Agrobacterium tumefaciens virB operon: virB2 through
virB11 are essential virulence genes. J Bacteriol 176, 3646–3660.
Buchanan-Wollaston, V., Passiatore, J. E. & Cannon, F. (1987). Themob and oriT mobilization functions of a bacterial plasmid promote
its transfer to plants. Nature 328, 172–175.
Cascales, E. & Christie, P. J. (2004a). Agrobacterium VirB10, an ATP
energy sensor required for type IV secretion. Proc Natl Acad Sci U S A
101, 17228–17233.
Cascales, E. & Christie, P. J. (2004b). Definition of a bacterial type IV
secretion pathway for a DNA substrate. Science 304, 1170–1173.
Chen, C. Y. & Kado, C. I. (1996). Osa protein encoded by plasmid pSa
is located at the inner membrane but does not inhibit membrane
association of VirB and VirD virulence proteins in Agrobacteriumtumefaciens. FEMS Microbiol Lett 135, 85–92.
Cho, H. & Winans, S. C. (2005). VirA and VirG activate the Ti plasmid
repABC operon, elevating plasmid copy number in response towound-released chemical signals. Proc Natl Acad Sci U S A 102,
14843–14848.
Christie, P. J. (2001). Type IV secretion: intercellular transfer of
macromolecules by systems ancestrally related to conjugation
machines. Mol Microbiol 40, 294–305.
Christie, P. J., Atmakuri, K., Krishnamoorthy, V., Jakubowski, S. &Cascales, E. (2005). Biogenesis, architecture, and function of
bacterial type iv secretion systems. Annu Rev Microbiol 59, 451–485.
Dellagostin, O. A., Esposito, G., Eales, L. J., Dale, J. W. & McFadden, J.(1995). Activity of mycobacterial promoters during intracellular and
extracellular growth. Microbiology 141, 1785–1792.
Fullner, K. J. & Nester, E. W. (1996). Temperature affects the T-DNA
transfer machinery of Agrobacterium tumefaciens. J Bacteriol 178,1498–1504.
Garfinkel, D. J., Simpson, R. B., Ream, L. W., White, F. F., Gordon, M. P.& Nester, E. W. (1981). Genetic analysis of crown gall: fine structuremap of the T-DNA by site-directed mutagenesis. Cell 27, 143–153.
Ilangumaran, S., Shankernarayan, N., Ramu, G. & Muthukkaruppan, V.(1994). Antibody response to recombinant 65-kDa, 70-kDa and 18-kDa
mycobacterial antigens in leprosy patients and healthy contacts in a
leprosy-endemic population. Int J Lepr Other Mycobact Dis 62, 245–255.
Judd, P. K., Kumar, R. B. & Das, A. (2005). Spatial location and
requirements for the assembly of the Agrobacterium tumefaciens type
IV secretion apparatus. Proc Natl Acad Sci U S A 102, 11498–11503.
Kado, C. I. & Heskett, M. G. (1970). Selective media for isolation of
Agrobacterium, Corynebacterium, Erwinia, Pseudomonas, and
Xanthomonas. Phytopathology 60, 969–976.
Kao, J. C., Perry, K. L. & Kado, C. I. (1982). Indoleacetic acid
complementation and its relation to host range specifying genes on
the Ti plasmid of Agrobacterium tumefaciens. Mol Gen Genet 188,
425–432.
Karunakaran, R., Mauchline, T. H., Hosie, A. H. & Poole, P. S. (2005).A family of promoter probe vectors incorporating autofluorescent
and chromogenic reporter proteins for studying gene expression in
Gram-negative bacteria. Microbiology 151, 3249–3256.
Krall, L., Wiedemann, U., Unsin, G., Weiss, S., Domke, N. & Baron, C.(2002). Detergent extraction identifies different VirB protein
subassemblies of the type IV secretion machinery in the membranes
of Agrobacterium tumefaciens. Proc Natl Acad Sci U S A 99, 11405–
11410.
Labes, M., Puhler, A. & Simon, R. (1990). A new family of RSF1010-
derived expression and lac-fusion broad-host-range vectors for gram-
negative bacteria. Gene 89, 37–46.
Lai, E. M. & Kado, C. I. (1998). Processed VirB2 is the major subunit of
the promiscuous pilus of Agrobacterium tumefaciens. J Bacteriol 180,
2711–2717.
Lai, E. M. & Kado, C. I. (2000). The T-pilus of Agrobacterium
tumefaciens. Trends Microbiol 8, 361–369.
Lai, E. M., Shih, H. W., Wen, S. R., Cheng, M. W., Hwang, H. H. & Chiu,S. H. (2006). Proteomic analysis of Agrobacterium tumefaciens
response to the vir gene inducer acetosyringone. Proteomics 6,
4130–4136.
Lau-Wong, I. C., Locke, T., Ellison, M. J., Raivio, T. L. & Frost, L. S.(2008). Activation of the Cpx regulon destabilizes the F plasmid
transfer activator, TraJ, via the HslVU protease in Escherichia coli. Mol
Microbiol 67, 516–527.
Lee, L. Y., Humara, J. M. & Gelvin, S. B. (2001). Novel constructions
to enable the integration of genes into the Agrobacterium tumefaciens
C58 genome. Mol Plant-Microbe Interact 14, 577–579.
Li, P. L., Everhart, D. M. & Farrand, S. K. (1998). Genetic and sequence
analysis of the pTiC58 trb locus, encoding a mating-pair formation
system related to members of the type IV secretion family. J Bacteriol
180, 6164–6172.
Lin, Y. Z., Chen, H. Y., Kao, R., Chang, S. P., Chang, S. J. & Lai, E. M.(2008). Proteomic analysis of rice defense response induced by
probenazole. Phytochemistry 69, 715–728.
Lini, N., Rehna, E. A., Shiburaj, S., Maheshwari, J. J., Shankernarayan,N. P. & Dharmalingam, K. (2008). Functional characterization of a
small heat shock protein from Mycobacterium leprae. BMC Microbiol 8,
208.
Liu, Z. & Binns, A. N. (2003). Functional subsets of the VirB type IV
transport complex proteins involved in the capacity of Agrobacterium
tumefaciens to serve as a recipient in virB-mediated conjugal transfer
of plasmid RSF1010. J Bacteriol 185, 3259–3269.
Liu, A. C., Shih, H. W., Hsu, T. & Lai, E. M. (2008). A citrate-inducible
gene, encoding a putative tricarboxylate transporter, is downregulated
by the organic solvent DMSO in Agrobacterium tumefaciens. J Appl
Microbiol 105, 1372–1383.
VirB-induced small heat-shock protein HspL
61
Lundquist, R. C., Close, T. J. & Kado, C. I. (1984). Geneticcomplementation of Agrobacterium tumefaciens Ti plasmid mutantsin the virulence region. Mol Gen Genet 193, 1–7.
Matuszewska, M., Kuczynska-Wisnik, D., Laskowska, E. & Liberek, K.(2005). The small heat shock protein IbpA of Escherichia coli cooperateswith IbpB in stabilization of thermally aggregated proteins in adisaggregation competent state. J Biol Chem 280, 12292–12298.
McCullen, C. A. & Binns, A. N. (2006). Agrobacterium tumefaciens andplant cell interactions and activities required for interkingdommacromolecular transfer. Annu Rev Cell Dev Biol 22, 101–127.
Munchbach, M., Dainese, P., Staudenmann, W., Narberhaus, F. &James, P. (1999a). Proteome analysis of heat shock proteinexpression in Bradyrhizobium japonicum. Eur J Biochem 264, 39–48.
Munchbach, M., Nocker, A. & Narberhaus, F. (1999b). Multiple smallheat shock proteins in rhizobia. J Bacteriol 181, 83–90.
Narberhaus, F. (2002). Alpha-crystallin-type heat shock proteins:socializing minichaperones in the context of a multichaperonenetwork. Microbiol Mol Biol Rev 66, 64–93.
Nevesinjac, A. Z. & Raivio, T. L. (2005). The Cpx envelope stressresponse affects expression of the type IV bundle-forming pili ofenteropathogenic Escherichia coli. J Bacteriol 187, 672–686.
Quandt, J. & Hynes, M. F. (1993). Versatile suicide vectors whichallow direct selection for gene replacement in gram-negative bacteria.Gene 127, 15–21.
Raivio, T. L. (2005). Envelope stress responses and Gram-negativebacterial pathogenesis. Mol Microbiol 56, 1119–1128.
Rashkova, S., Zhou, X. R., Chen, J. & Christie, P. J. (2000). Self-assembly of the Agrobacterium tumefaciens VirB11 traffic ATPase.J Bacteriol 182, 4137–4145.
Rogowsky, P. M., Close, T. J., Chimera, J. A., Shaw, J. J. & Kado, C. I.(1987). Regulation of the vir genes of Agrobacterium tumefaciensplasmid pTiC58. J Bacteriol 169, 5101–5112.
Rosen, R., Buttner, K., Becher, D., Nakahigashi, K., Yura, T.,Hecker, M. & Ron, E. Z. (2002). Heat shock proteome ofAgrobacterium tumefaciens: evidence for new control systems.J Bacteriol 184, 1772–1778.
Rowley, G., Spector, M., Kormanec, J. & Roberts, M. (2006). Pushingthe envelope: extracytoplasmic stress responses in bacterial pathogens.Nat Rev Microbiol 4, 383–394.
Sambrook, J. & Russell, D. (2001). Molecular Cloning: a LaboratoryManual, 3rd edn. Cold Spring Harbor, NY: Cold Spring HarborLaboratory.
Schagger, H. & von Jagow, G. (1987). Tricine-sodium dodecylsulfate-polyacrylamide gel electrophoresis for the separation ofproteins in the range from 1 to 100 kDa. Anal Biochem 166, 368–379.
Schweizer, H. D. (1993). Small broad-host-range gentamycinresistance gene cassettes for site-specific insertion and deletionmutagenesis. Biotechniques 15, 831–834.
Shirasu, K. & Kado, C. I. (1993). Membrane location of the Ti plasmidVirB proteins involved in the biosynthesis of a pilin-like conjugativestructure on Agrobacterium tumefaciens. FEMS Microbiol Lett 111,287–294.
Shurvinton, C. E. & Ream, W. (1991). Stimulation of Agrobacteriumtumefaciens T-DNA transfer by overdrive depends on a flankingsequence but not on helical position with respect to the border repeat.J Bacteriol 173, 5558–5563.
Simon, R., Priefer, U. & Puhler, A. (1983). A broad host rangemobilization system for in vivo genetic engineering: transposonmutagenesis in gram negative bacteria. Biotechnology 1, 784–791.
Stewart, G. R., Newton, S. M., Wilkinson, K. A., Humphreys, I. R.,Murphy, H. N., Robertson, B. D., Wilkinson, R. J. & Young, D. B.(2005). The stress-responsive chaperone alpha-crystallin 2 is requiredfor pathogenesis of Mycobacterium tuberculosis. Mol Microbiol 55,1127–1137.
Stewart, J. N., Rivera, H. N., Karls, R., Quinn, F. D., Roman, J. &Rivera-Marrero, C. A. (2006). Increased pathology in lungs of miceafter infection with an alpha-crystallin mutant of Mycobacteriumtuberculosis: changes in cathepsin proteases and certain cytokines.Microbiology 152, 233–244.
Studer, S. & Narberhaus, F. (2000). Chaperone activity and homo-and hetero-oligomer formation of bacterial small heat shock proteins.J Biol Chem 275, 37212–37218.
Sun, Y. & MacRae, T. H. (2005). Small heat shock proteins: molecularstructure and chaperone function. Cell Mol Life Sci 62, 2460–2476.
von Bodman, S. B., McCutchan, J. E. & Farrand, S. K. (1989).Characterization of conjugal transfer functions of Agrobacteriumtumefaciens Ti plasmid pTiC58. J Bacteriol 171, 5281–5289.
Ward, D. V., Draper, O., Zupan, J. R. & Zambryski, P. C. (2002).Peptide linkage mapping of the Agrobacterium tumefaciens vir-encoded type IV secretion system reveals protein subassemblies.Proc Natl Acad Sci U S A 99, 11493–11500.
Watson, B., Currier, T. C., Gordon, M. P., Chilton, M. D. & Nester, E. W.(1975). Plasmid required for virulence of Agrobacterium tumefaciens.J Bacteriol 123, 255–264.
Wilkinson, K. A., Stewart, G. R., Newton, S. M., Vordermeier, H. M.,Wain, J. R., Murphy, H. N., Horner, K., Young, D. B. & Wilkinson, R. J.(2005). Infection biology of a novel alpha-crystallin of Mycobacteriumtuberculosis: Acr2. J Immunol 174, 4237–4243.
Wu, H. Y., Chung, P. C., Shih, H. W., Wen, S. R. & Lai, E. M. (2008).Secretome analysis uncovers an Hcp-family protein secreted via a typeVI secretion system in Agrobacterium tumefaciens. J Bacteriol 190,2841–2850.
Yuan, Q., Carle, A., Gao, C., Sivanesan, D., Aly, K. A., Hoppner, C.,Krall, L., Domke, N. & Baron, C. (2005). Identification of the VirB4-VirB8-VirB5-VirB2 pilus assembly sequence of type IV secretionsystems. J Biol Chem 280, 26349–26359.
Zahrl, D., Wagner, M., Bischof, K. & Koraimann, G. (2006). Expressionand assembly of a functional type IV secretion system elicitextracytoplasmic and cytoplasmic stress responses in Escherichia coli.J Bacteriol 188, 6611–6621.
Zuber, P. & Losick, R. (1983). Use of a lacZ fusion to study the role ofthe spoO genes of Bacillus subtilis in developmental regulation. Cell35, 275–283.
Edited by: C. A. Boucher
Y.-L. Tsai and others
62
Diskussion
63
E Diskussion
1. Biosynthese von Phosphatidylcholin in A. tumefaciens
PC galt lange Zeit als ausschließlich eukaryotisches Membranlipid. Jedoch zeigten
Genomsequenzanalysen bekannter Bakterienarten, dass etwa 10 % aller bisher
identifizierten Bakterien grundsätzlich zur PC-Biosynthese befähigt sind
(Sohlenkamp et al., 2003). In A. tumefaciens wurden anhand der Genomsequenz
beide in Bakterien bekannten PC-Biosynthesewege vorhergesagt. Für die drei
Reaktionen des Methylierungsweges wurde eine Phospholipid N-Methyltransferase
(PmtA) und für die Katalyse der direkten Kondensation von Cholin mit CDP-
Diacylglycerin eine PC-Synthase (Pcs) postuliert. Durch die Charakterisierung des
Lipidspektrums verschiedener A. tumefaciens PC-Biosynthesemutanten (Wessel et
al., 2006) sowie durch heterologe Expression beider Enzyme in E. coli wurde die
annotierte Funktion für PmtA und Pcs eindeutig bestätigt.
In Prokaryoten ist bisher über mögliche Regulationsmechanismen der PC-
Biosynthese wenig bekannt. Reportergenstudien und RT-PCR-Analysen der pmtA-
und pcs-Gene aus A. tumefaciens zeigten, dass die Transkription beider PC-
Biosynthesegene konstitutiv von einem eigenen Promoter erfolgt.
Transkriptomstudien mit A. tumefaciens unter verschiedenen Bedingungen liefern
keine Hinweise auf eine mögliche Regulation der pmtA- und pcs-Genexpression
durch Azidität, verschiedene Pflanzensignale (γ-Aminobuttersäure, Salicylsäure und
Indol-3-Essigsäure) oder dem Virulenzinduktor Acetosyringon (Anand et al., 2008;
Yuan et al., 2008a; Yuan et al., 2008b). Durch diese Studien kann eine Regulation
der PC-Biosynthese auf transkriptioneller Ebene für die getesteten Bedingungen
ausgeschlossen werden.
Im Gegensatz zu A. tumefaciens wurden in B. japonicum Hinweise für eine
transkriptionelle Regulation einiger PC-Biosynthesegene beschrieben. Grundsätzlich
ist in B. japonicum die PC-Biosynthese durch das Vorkommen vier verschiedener
pmt-Gene und eines pcs-Gens komplex. Im Wildtyp werden unter aeroben und
anaeroben Bedingungen pmtA, pmtX1 und pcs exprimiert, nicht aber pmtX2-4
(Hacker et al., 2008b). Interessanterweise wurde in einer pmtA-Mutante, die nur noch
eine reduzierte PC-Menge in den Membranen besitzt, pmtX1, pmtX4 und pcs
Diskussion
64
exprimiert. Durch die Induktion der pmtX4-Expression scheint der Verlust von PmtA
zumindest teilweise kompensiert zu werden, um weiterhin die PC-Synthese zu
gewährleisten (Hacker et al., 2008a). Durch vergleichende Transkriptomstudien mit
B. japonicum Wildtyp und einer irr-Mutante wurde gezeigt, dass pmtX1 in der irr-
Mutante unter Eisen-limitierenden Bedingungen dreifach reprimiert wird (Yang et al.,
2006). Irr ist ein Regulatorprotein, das unter Eisen-limitierenden Bedingungen
akkumuliert und den Eisenmetabolismus kontrolliert. Des Weiteren wird die pmtA-
Expression unter Hitzeschockbedingungen um einen Faktor 4 reprimiert (Minder et
al., 2001). Die Expressionsstudien geben erste Hinweise darauf, dass die PC-
Biosynthese in B. japonicum unter bestimmten Stressbedingungen auf
transkriptioneller Ebene reguliert wird. Möglicherweise führt dies zu einer dem
jeweiligen Stress angepassten Membranzusammensetzung, die dem Organismus
optimalen Schutz vor äußeren Umwelteinflüssen bietet.
Für Säugetiere ist bekannt, dass die PC-Biosynthese auf trankriptioneller Ebene
reguliert wird. Auch die Enzymaktivitäten spielen eine entscheidende Rolle (Kent,
1990; Li & Vance, 2008; Ridgway & Vance, 1988; Ridgway et al., 1989). Da die PC-
Biosynthese in A. tumefaciens keiner transkriptionalen Regulation zu unterliegen
scheint, könnte diese durch die Aktivitäten der Enzyme PmtA und Pcs kontrolliert
werden. Eine biochemische Charakterisierung des PmtA-Enzyms zeigte, dass die
Aktivität durch die Produkte PC und SAH inhibiert wird (Aktas & Narberhaus, 2009).
Einen weiteren Hinweis für die Regulation auf Enzym-Ebene liefert die Lipidanalyse
der A. tumefaciens pmtA-Mutante, die in Minimalmedium mit verschiedenen
Cholinkonzentrationen kultiviert wurde. Mit steigender Cholinkonzentration erhöhte
sich auch die über den PC-Synthaseweg gebildete PC-Menge (Daten nicht gezeigt,
Sonja Klüsener). Da die Reportergenstudien eine Regulation auf transkriptioneller
Ebene durch Zugabe von Cholin ausschließen, könnte die Verfügbarkeit des
Substrates Cholin die Aktivität von Pcs stimulieren.
Gemäß der Nutzung und Gewichtung der beiden PC-Biosynthesewege können
Prokaryoten in drei Gruppen eingeteilt werden (Tab. 2): i) A. aceti, Z. mobilis und
R. sphaeroides interagieren mit keinem eukaryotischen Wirt und nutzen nur den
Methylierungsweg zur PC-Biosynthese (Arondel et al., 1993; Hanada et al., 2001;
Martínez-Morales et al., 2003; Tahara et al., 1994). ii) Die tierpathogenen Bakterien
B. abortus, B. melitensis, B. burgdorferi, P. aeruginosa und L. pneumophila bilden
PC vorwiegend über den PC-Synthaseweg (Comerci et al., 2006; Conde-Alvarez et
Diskussion
65
al., 2006; Conover et al., 2008; Martínez-Morales et al., 2003; Wilderman et al.,
2002). iii) Die Pflanzensymbionten B. japonicum, S. meliloti,
Rhizobium leguminosarum und Mesorhizobium loti produzieren PC über beide
Synthesewege (de Rudder et al., 1999; de Rudder et al., 1997; Hacker et al., 2008b;
Martínez-Morales et al., 2003; Minder et al., 2001).
Tab. 2: Phosphatidylcholin-Biosynthese in Prokaryoten. Die Charakterisierung beruht auf Aktivitätsversuchen (PmtA und Pcs) mit Zellextrakten und/oder in vivo Experimenten mit PC-Biosynthesemutanten.
Interaktion PC-Synthesewege Organismus Wirt-Bakterium PmtA Pcs
Acetobacter aceti keine ja ---
Zymomonas mobilis keine ja ---
Rhodobacter sphaeroides keine ja ---
Brucella abortus Tierpathogen (ja) ja
Brucella melitensis Tierpathogen --- ja
Borrelia burgdorferi Tierpathogen --- ja
Pseudomonas aeruginosa Tierpathogen --- ja
Legionella pneumophila Tierpathogen (ja) ja
Agrobacterium tumefaciens Pflanzenpathogen ja ja
Bradyrhizobium japonicum Pflanzensymbiont ja ja
Sinorhizobium meliloti Pflanzensymbiont ja ja
Rhizobium leguminosarum Pflanzensymbiont ja ja
Mesorhizobium loti Pflanzensymbiont ja ja
(ja), Enzymaktivität vorhanden, aber entsprechender PC-Biosyntheseweg spielt in vivo nur eine untergeordenete Rolle; ---, keine Enzymaktivität vorhanden.
Die Varianz in der Nutzung der PC-Biosynthesewege könnte mit der Verfügbarkeit
von Cholin für den PC-Synthaseweg zusammenhängen. Generell ist der PC-
Synthaseweg im Gegensatz zu dem Methylierungsweg für das Bakterium
energetisch günstiger (Sohlenkamp et al., 2003). Für die tierpathogenen Bakterien
stehen während der Infektion durch den Wirt große Mengen an Cholin zur
Verfügung, so dass diese den PC-Synthaseweg nutzen können. Obwohl für
B. abortus und L. pneumophila Pmt-Aktivitäten nachgewiesen wurden, spielt der
Methylierungsweg jedoch in vivo nur eine untergeordnete Rolle (Comerci et al., 2006;
Diskussion
66
Conde-Alvarez et al., 2006; Conover et al., 2008; Martínez-Morales et al., 2003).
Bakterien, die nicht mit einem eukaryotischen Wirt interagieren, müssten Cholin
entweder de novo synthetisieren oder aus der Umgebung importieren. Nicht alle
Bakterien können Cholin herstellen und es ist auch nicht immer in der freien
Umgebung verfügbar, so dass die PC-Biosynthese in diesen Bakterien durch die
Nutzung des Methylierungsweges sichergestellt werden muss. Einer erfolgreichen
Symbiose liegt eine komplexe Kommunikation zwischen dem Wirt und dem
Bakterium zugrunde, die dazu führt, dass beide Seiten einen optimalen Nutzen von
der Interaktion haben. Es könnte sein, dass die verfügbare Cholin-Menge in der
Pflanze die Gewichtung der PC-Biosynthesewege der Pflanzensymbionten bestimmt.
Dies würde bedeuten, dass bei ausreichender Cholin-Menge für beide Partner der
Symbiose der PC-Synthaseweg genutzt wird und bei niedriger Cholin-Menge der
Methylierungsweg.
Da A. tumefaciens ein Pflanzenpathogen ist, könnte es in Bezug auf die Gewichtung
der PC-Biosynthesewege zwischen den Tierpathogenen und den
Pflanzensymbionten eingeordnet werden. A. tumefaciens besitzt einerseits die
Möglichkeit, PC über beide Wege zu produzieren. Andererseits muss es aber auch
keine Rücksicht auf den Interaktionspartner nehmen und kann das verfügbare Cholin
für den PC-Synthaseweg nutzen. Aus den bisher bekannten Daten, kann folgendes
Modell zur PC-Biosynthese in A. tumefaciens postuliert werden (Abb. 10). In
Anwesenheit von Cholin wird die Aktivität des Pcs-Enzyms stimuliert, so dass PC
über den PC-Synthaseweg durch direkte Kondensation von CDP-Diacylglycerin und
Cholin in der inneren Membran gebildet wird. Die produzierten PC-Moleküle hemmen
die PmtA-Aktivität, um die Nutzung des im Gegensatz zum Methylierungsweg
energetisch günstigeren PC-Synthaseweges zu gewährleisten. PC verbreitet sich
durch laterale Diffusion und Flip-Flop-Mechanismen in der Lipiddoppelschicht und
wird teilweise aktiv in die äußere Membran transportiert (Doerrler, 2006). Wenn
Cholin nur in sehr geringen Mengen oder gar nicht zur Verfügung steht, sinkt die
Pcs-Aktivität, so dass nur eine geringe PC-Menge über den PC-Synthaseweg
gebildet wird. Dadurch wird PmtA nicht inhibiert und kann PE über die Intermediate
MMPE und DMPE zu PC methylieren. Hierbei wird die PC-Biosynthese über die
Hemmung der PmtA-Aktivität durch die Produkte PC und SAH reguliert (Aktas &
Narberhaus, 2009).
Diskussion
67
Abb. 10: Modell für die Phosphatidylcholin-Biosynthese in A. tumefaciens. Wenn Cholin für die Zelle verfügbar ist, wird PC durch die Verbindung von CDP-Diacylglycerin und Cholin mittels der PC-Synthase (Pcs) gebildet. Ob internes oder externes Cholin umgesetzt wird, ist nicht bekannt. Die synthetisierten PC-Moleküle hemmen die Aktivität der Phospholipid N-Methyltransferase (PmtA). Die Lipide bewegen sich durch den Flip-Flop Mechanismus und der lateralen Diffusion innerhalb der Lipiddoppelschicht und werden in die äußere Membran transportiert. CDP, Cytidindiphosphat; CMP, Cytidinmonophosphat.
2. Globale Auswirkungen der Phosphatidylcholin-Defizienz in A. tumefaciens
In der PC-defizienten A. tumefaciens Mutante sind im Vergleich zum Wildtyp bis zu
13 % aller Gene unterschiedlich exprimiert. Diese kodieren hauptsächlich für
Membranproteine oder für Proteine, die in Membran-assoziierten Prozessen beteiligt
sind. Ebenso zeigt das Transkriptom einer B. japonicum pmtA-Mutante
unterschiedlich exprimierte Gene, die hauptsächlich für Membranproteine kodieren
oder für Proteine, die mit den Membranen funktionell verbunden sind (Hacker et al.,
2008a). In beiden Organismen sind in den PC-Biosynthesemutanten z.B. Zwei-
Komponentensysteme oder Transporter im Vergleich zum Wildtyp unterschiedlich
exprimiert. Die B. japonicum pmtA-Mutante weist im Gegensatz zu der PC-
defizienten A. tumefaciens Mutante noch eine reduzierte PC-Menge auf, wodurch die
Expression einer nur limitierten Anzahl von Genen in der bradyrhizobiellen pmtA-
Mutante erklärt werden kann (Hacker et al., 2008a).
Da die PC-defiziente A. tumefaciens Mutante sowohl in Voll- als auch in
Minimalmedium vergleichbar zum Wildtyp wächst, scheint die Mutante den
Membranstress erfolgreich durch einen veränderten Proteinpool, insbesondere der
Membranproteine, zu kompensieren. Dies wird beispielhaft an der stark veränderten
Expression vieler Gene deutlich, deren Genprodukte im Eisenmetabolismus involviert
sind. Die signifikante Induktion von Hämrezeptoren, Eisen-Transportern und Eisen-
Diskussion
68
Chelatoren gewährleistet trotz der veränderten Membran den Eisen-Import, um das
für A. tumefaciens essentielle Metall in Wildtyp-ähnlichen Mengen für zelluläre
Funktionen zur Verfügung zu stellen. Die phänotypische Charakterisierung
verschiedener B. subtilis Stämme mit stark veränderter Membranzusammensetzung
(CL-, PE-, Lyso-PG- oder Glykolipid-Mutanten) zeigen ebenfalls ein normales
Wachstumsverhalten in Voll- und Minimalmedium sowie keine Veränderungen in der
Sporulation (Salzberg & Helmann, 2008). Ebenso wie in der PC-defizienten
A. tumefaciens-Mutante ist auf transkriptioneller Ebene jedoch eine hohe Anzahl von
Genen verändert, die u.a. für ABC-Transporter oder Efflux-Pumpen kodieren sowie
für Proteine, die in der Zellwand- und Membran-Biosynthese oder in der Motilität
involviert sind.
Eine PC-defiziente S. meliloti-Mutante hingegen weist im Vergleich zum Wildtyp ein
deutlich verlangsamtes Wachstum auf. Das deutet darauf hin, dass die
Membranveränderungen in diesem Organismus nicht ausreichend kompensiert
werden (de Rudder et al., 2000). Vergleicht man die PC-Mengen von A. tumefaciens
und S. meliloti miteinander, so liegt die PC-Menge mit 36,5 % in den Membranen von
S. meliloti (Sohlenkamp et al., 2000) über dem prozentualen PC-Anteil in den
Membranen von A. tumefaciens (~ 23 %). Die Lipidzusammensetzung der PC-
defizienten Membranen von S. meliloti scheint somit stärker gestört zu sein als in
A. tumefaciens, was das verlangsamte Wachstum erklären könnte.
Membranen dienen allen Zellen als Schutz und Permeabilitätsbarriere und sind somit
ein essentieller Baustein des Lebens. In Bakterien beeinflusst die Membran-
zusammensetzung durch die physikalischen und chemischen Eigenschaften der
Lipide viele zelluläre Prozesse wie die Zellteilung, die DNA-Replikation oder den
Proteintransport (Cronan, 2003; Ichihashi et al., 2003; Norris & Fishov, 2001). Die
Phospholipideigenschaften sind sowohl von der Kopfgruppe abhängig als auch von
der Länge und dem Sättigungsgrad der Fettsäurereste. Ungesättigte Fettsäuren
besitzen einen niedrigeren Schmelzpunkt als gesättigte Fettsäuren und Fettsäuren
mit kurzen Ketten weisen wiederum einen niedrigeren Schmelzpunkt auf als
langkettige Fettsäuren (Cronan & Gelmann, 1975). Die Regulation der Fettsäure-
Biosynthese ist für die Membran-Homöostase essentiell, da die biophysikalischen
Eigenschaften der Membranen in besonderem Maße durch die Fettsäure-
zusammensetzung bestimmt werden. Aufgrund ihrer Geometrie sind Phospholipide
mit ungesättigten Fettsäuren weniger dicht angeordnet als Lipide mit gesättigten
Diskussion
69
Fettsäuren, wodurch diese Membranbereiche fluider sind als andere (Cronan &
Gelmann, 1975) (Abb. 11). Die Fluidität der Membranen ist u.a. wichtig für die
laterale Diffusion der Membranproteine und somit für die Assemblierung
verschiedener Proteinkomplexe, die in den Membranen lokalisiert oder funktionell an
den Membranen assoziiert sind.
Abb. 11: Biophysikalische Eigen-schaften der Fettsäuren. Die Fluidität der Lipiddoppelschichten wird erheblich durch den Sättigungs-grad und die Geometrie der Fett-säuren bestimmt. Bei ungesättigten Fettsäuren mit cis-Bindung liegen die Wasserstoffatome an der Doppel-bindung auf der gleichen Seite vor, so dass ein 30°-Knick entsteht. Bei einer trans-Bindung befinden sich die Wasserstoffatome diagonal gegen-über, wodurch die Geometrie der Fettsäuren linearer ist.
Obwohl die biophysikalischen Eigenschaften der Membranen ebenfalls durch
Modifikationen an der Kopfgruppe der Phospholipide beeinflusst werden können,
nutzen Bakterien biochemische und genetische Mechanismen, um die de novo
Fettsäure-Biosynthese an sich ändernde Umwelteinflüsse anzupassen (Cronan &
Gelmann, 1975; Zhang & Rock, 2008). In B. subtilis wird die Fettsäure-Biosynthese
und der Sättigungsgrad der Fettsäuren durch Osmolarität und Ethanol reguliert
(López et al., 1998; Rigomier et al., 1980). Setzt man E. coli-Zellen Ethanolstress
aus, so verändert sich das Verhältnis von PE und PG nicht, jedoch ist ein erheblicher
Anstieg der einfach ungesättigten Fettsäure Vaccensäure zu beobachten (Ingram,
1977). Des Weiteren nimmt die Anzahl an gesättigten Fettsäuren in den Membranen
vieler Bakterien bei steigender Wachstumstemperatur zu (Marr & Ingraham, 1962).
Nicht nur Umwelteinflüsse können den Fettsäure-Metabolismus beeinflussen,
sondern auch direkter Membranstress durch Veränderungen in der Phospholipid-
zusammensetzung. In der Hefe S. cerevisiae führt die Reduktion der PC-Menge zum
verstärkten Einbau von verkürzten und gesättigten Fettsäuren (de Kroon, 2007).
A. tumefaciens kompensiert die Auswirkungen der PC-defizienten Membranen durch
Umgestaltung des Proteinpools und scheint dabei über den Fettsäure-Metabolismus
die veränderten biophysikalischen Eigenschaften der Membranen auszugleichen.
Viele Gene, die eine zentrale Rolle in der Initiation und Elongation der Fettsäure-
Diskussion
70
Biosynthese spielen, sind in der Mutante induziert. Hingegen werden Gene, die im
Fettsäure-Abbau beteiligt sind, reprimiert (Tab. 3). Ob die Fettsäure-
zusammensetzung in der PC-defizienten Mutante gegenüber dem Wildtyp verändert
ist, soll in Zukunft geklärt werden.
Tab. 3: Veränderungen im Fettsäure-Metabolismus in Abwesenheit von Phosphatidylcholin. Zusammenfassung der unterschiedlich exprimierten Gene von A. tumefaciens Wildtyp und PC-defizienter Mutante, die für im Fettsäure-Metabolismus involvierte Enzyme kodieren.
Fettsäure-Metabolismusa ∆pmtA ∆pcs vs. WTb Gen-Nr. Gen-Name Stoffwechselwege -AS +AS
atu0757 fabI Fettsäure-Elongation - gesättigt 4 4
atu1097 fabF Fettsäure-Elongation - gesättigt
Fettsäure-Biosynthese - Initiation
2 -
atu1179 fabH Fettsäure-Elongation - gesättigt
Fettsäure-Biosynthese - Initiation
4 3
atu1331 accB Fettsäure-Biosynthese - Initiation 3 3
atu1599 fabZ Fettsäure-Elongation - gesättigt 3 3
atu3505 --- β-Oxidation 2 -
atu3813 fabG Fettsäure-Elongation - gesättigt 3 2
atu3861 --- Acyl-CoA Hydrolyse -2 -
atu4001 fabH Fettsäure-Elongation - gesättigt
Fettsäure-Biosynthese - Initiation
- 2
atu5122 fabG Fettsäure-Elongation - gesättigt - 2
atu6066 acd β-Oxidation -5 -11 a Informationen veröffentlicht durch (Goodner et al., 2001; Wood et al., 2001). b Die Genexpression wurde nach einem Vergleich der beiden Stämme als unterschiedlich definiert, wenn ein Fold Change von >2 vorlag; -, Genexpression unterscheidet sich nicht; AS, Virulenzinduktor Acetosyringon.
Interessanterweise ergaben die vergleichenden Transkriptomstudien mit
A. tumefaciens Wildtyp und PC-defizienter Mutante keine Veränderungen in der
Expression von Genen, die in der Phospholipid-Biosynthese involviert sind (außer
pmtA und pcs) und somit Variationen in der Kopfgruppe hervorrufen würden.
PC besitzt durch die große Kopfgruppe eine zylinderförmige Geometrie, welche die
Ausbildung von Lipiddoppelschichten begünstigt. Das Vorläufermolekül PE hingegen
hat durch die kleine Kopfgruppe eine kegelförmige Geometrie, die der Ausbildung
Diskussion
71
von Doppelschichten entgegen wirkt (Boumann et al., 2006; de Kroon, 2007). Fehlt
PC in den Membranen, müssen die biophysikalischen Eigenschaften der PE-
Moleküle so geändert werden, dass die Ausbildung einer funktionsfähigen
Lipiddoppelschicht mit geeigneter Fluidität weiterhin gewährleistet ist. Der Einbau
von gesättigten Fettsäuren würde hierbei die Stabilität der Lipiddoppelschicht fördern
(Boumann et al., 2006; de Kroon, 2007) (Abb. 11). Betrachtet man die in der PC-
defizienten A. tumefaciens-Mutante unterschiedlich exprimierten Gene aus Tabelle 3,
so kodieren viele für Schlüsselenzyme in der Biosynthese für gesättigte Fettsäuren
wie fabI, fabF oder fabZ. Demnach reguliert A. tumefaciens das Gleichgewicht der
biophysikalischen Eigenschaften der verschiedenen Membrankomponenten
vermutlich über den Fettsäure-Metabolismus, um eine funktionsfähige Permeabilität
und Fluidität sicher zu stellen.
3. Einfluss von Phosphatidylcholin auf die Biofilmbildung und Motilität
A. tumefaciens-Zellen mit PC-freien Membranen weisen nach Kultivierung in
Minimalmedium einen signifikanten Motilitätsdefekt und eine veränderte
Biofilmbildung auf. Das Flagellen-abhängige Schwimmverhalten der PC-defizienten
Mutante ist im Vergleich zum Wildtyp stark reduziert, was auf eine geringere Menge
der für die Flagellen-Struktur essentiellen FlaA- und FlaB-Proteine zurückzuführen
ist. Da die Transkription der entsprechenden Gene nicht reduziert ist, müssen die
geringen FlaA- und FlaB-Mengen posttranskriptionell begründet sein. Für eine
erfolgreiche Adhäsion und Biofilmbildung von A. tumefaciens-Zellen an biotischen
und abiotischen Oberflächen spielen verschiedene Faktoren eine wichtige Rolle. Die
Produktion sowie Sekretion von β-(1,2)-Glukanen und Cellulose sind für die
Adhäsion und Biofilmbildung essentiell (de Iannino & Ugalde, 1989; Douglas et al.,
1985; Matthysse, 1983; Matthysse et al., 2005). In der PC-defizienten Mutante
werden weder die Gene für die β-(1,2)-Glukan-Synthese (chvA und chvB) noch die
der Cellulose-Synthese (celI, celABC, celG und celED) im Vergleich zum Wildtyp
unterschiedlich exprimiert, so dass die veränderte Biofilmbildung der Mutante nicht
darauf zurückzuführen ist. Allerdings zeigen die vergleichenden
Transkriptomanalysen eine Reduktion von sinR um einen Faktor von 5 unter nicht-
induzierten bzw. 7 unter Virulenz-induzierten Bedingungen. Der Transkriptions-
Diskussion
72
aktivator SinR ist als Regulator der Biofilmbildung beschrieben und wichtig für die
Entwicklung des Biofilms, wobei die Zielgene unbekannt sind. Eine sinR-Mutante
weist eine drastisch verminderte Biofilmbildung auf (Ramey et al., 2004). Die PC-
defiziente A. tumefaciens-Mutante aggregiert im Vergleich zum Wildtyp früher und
bildet mehr Biomasse in drei-dimensionalen Strukturen aus. Damit steht der
Phänotyp der PC-defizienten Mutante im Gegensatz zu der sinR-Repression, so
dass auch für die veränderte Biofilmbildung posttranskriptionelle Gründe
wahrscheinlich sind.
Sowohl Motilität als auch Biofilmbildung sind komplexe Prozesse, die über die
Membranen vermittelt werden und oft miteinander verknüpft sind. So weisen
A. tumefaciens-Mutanten mit einem drastischen Motilitätsdefekt (ΔflgE, Δmot,
ΔflaABC) ebenfalls eine stark verminderte Adhäsion und Biofilmbildung auf
(Chesnokova et al., 1997; Merritt et al., 2007). In der PC-defizienten A. tumefaciens-
Mutante werden die Veränderungen in der Membranzusammensetzung über die
Anpassung des Proteinpools und des Fettsäure-Metabolismus kompensiert, um ein
normales Wachstum zu gewährleisten (Kapitel 2). Trotzdem weisen die PC-freien
Membranen vermutlich nicht vollständig die biophysikalischen Eigenschaften der
Wildtyp Membranen auf. Dies könnte zur fehlerhaften Assemblierung Membran-
durchspannender Proteinkomplexe wie dem Flagellen-Apparat führen. E. coli PE-
Mutanten zeigen ein stark reduziertes Schwimmverhalten im Vergleich zum Wildtyp.
Der Motilitätsdefekt lässt sich durch eine geringere Flagellenanzahl erklären, die auf
eine reduzierte Transkription der entsprechenden Gene zurückzuführen ist. Die
Autoren vermuten hierbei eine fehlerhafte Assemblierung der Flagellen, wodurch die
Vorläuferproteine akkumulieren und die Transkription der Gene inhibieren (Shi et al.,
1993). Des Weiteren ist die PC-Menge in den Membranen von
Bradyrhizobium sp. SEMIA 6144 und L. pneumophila kritisch für das Flagellen-
abhängige Schwimmverhalten, wobei die jeweiligen Ursachen für den Motilitätsdefekt
unklar sind (Conover et al., 2008; Medeot et al., 2010).
A. tumefaciens besitzt für die Proteinsekretion über die Membranen neben den
Typ I-, Typ IV-, Typ V-, Sec- und Tat-Sekretionssystemen, ein funktionales Typ VI-
Sekretionssystem, dessen biologische Funktion noch ungeklärt ist (Ma et al., 2009;
Wu et al., 2008). Vergleichende Untersuchungen des TVISS in A. tumefaciens
Wildtyp und PC-defizienter Mutante liefern einen weiteren Hinweis für die fehlerhafte
Assemblierung Membran-durchspannender Proteinkomplexe in Abwesenheit von
Diskussion
73
PC. Mittels Western-Blot-Analysen konnten nach Anzucht in Minimalmedium
verschiedene Komponenten des TVISS in vergleichenden Proteinmengen detektiert
werden. Jedoch ist die sekretierte Proteinmenge des Substrats Hcp (Hemolysin-
coregulated protein) in der PC-defizienten Mutante deutlich geringer als im Wildtyp
(Daten nicht gezeigt, Sonja Klüsener). Dies bedeutet, dass die verschiedenen
Komponenten des TVISS in Wildtyp-ähnlichen Mengen produziert werden. Jedoch
scheint der Sekretionsapparat nicht vollständig funktional zu sein, da die Hcp-
Sekretion vermindert ist.
PC könnte in A. tumefaciens aber auch spezifisch an der Lokalisation oder
Funktionalität bestimmter Proteine beteiligt sein, die wiederum für die Motilität und
Biofilmbildung wichtig sind. In E. coli fungiert PE als Chaperon für die korrekte
Insertion der Laktose-Permease LacY in die innere Membran (Bogdanov et al.,
2002). Die anionischen Lipide PG und CL sind für die DNA-Replikation von enormer
Bedeutung, da sie den Zyklus von ADP-DnaA zu der aktiven Form ATP-DnaA am
oriC vermitteln (Crooke et al., 1992). In Synechocystis sp. PCC6803 ist die Aktivität
des Photosystems II während der Photosynthese direkt von den PG-Mengen in den
Thylakoidmembranen abhängig (Hagio et al., 2000). In E. coli-Mutanten mit
reduziertem PG-Gehalt ist die Translokation der Vorläuferproteine von PhoE und
OmpA über die innere Membran beeinträchtigt (de Vrije et al., 1988). Außerdem
wurde gezeigt, dass der PG-Gehalt in den Membranen von E. coli kritisch für ein
funktionelles Sec-Sekretionssystem ist. Hierbei ist die korrekte Insertion von SecA,
einem Schlüssenenzym des Sec-Sekretionssystems, in die innere Membran und
dessen ATPase-Aktivität PG-abhängig (Lill et al., 1990; Shinkai et al., 1991).
In A. tumefaciens könnte ein nicht funktionales Tat-Sekretionssystem in den PC-
defizienten Membranen die Ursache für den Motilitätsdefekt und die veränderte
Biofilmbildung sein. Das Tat-Sekretionssystem ist für die Translokation gefalteter
Proteine über die innere Membran zuständig und besteht aus einem TatBC-Komplex
und einem TatA-Komplex (Abb. 12). Der Transportzyklus startet, wenn ein Substrat
durch den TatBC-Komplex erkannt wird und der TatA-Komplex mit dem TatBC-
Komplex assoziiert. So kann dann das Substrat durch den Kanal des TatA-
Komplexes ins Periplasma transportiert werden. Nach erfolgreicher Translokation
dissoziieren die beiden Komplexe wieder (De Buck et al., 2008b).
In A. tumefaciens ist das Tat-Sekretionssystem für die Synthese und Funktion der
Flagelle wichtig (Ding & Christie, 2003). Weitere Prozesse wie Chemotaxis und
Diskussion
74
Virulenz sind ebenfalls in einer A. tumefaciens tatC-Mutante reduziert (Abb. 12). In
verschiedenen Bakterien wurde gezeigt, dass das Tat-Sekretionssystem viele
zelluläre Prozesse beeinflusst. L. pneumophila tat-Mutanten weisen eine veränderte
Biofilmbildung auf und haben geringere Flagellin-Mengen, welche für die Struktur der
Flagelle essentiell sind (De Buck et al., 2008a; De Buck et al., 2005). In
P. aeruginosa ist das Tat-Sekretionssystem in die Motilität, Biofilmbildung, Virulenz
und den Eisen-Metabolismus involviert. Die für die Virulenz wichtigen
Phospholipasen des Typs C (PlcN und PlcH) und das im Eisen-Metabolismus
wichtige Protein PvdN (Siderophor-Synthese) wurden als Tat-Substrate identifiziert
(Ochsner et al., 2002). Das Tat-Sekretionssystem hat in Yersinia pseudotuberculosis
Einfluss auf die Motilität, Biofilmbildung, Stressadaption und Virulenz (Lavander et
al., 2006). Interessanterweise konnten für keinen der aufgeführten Organismen Tat-
Substrate identifiziert werden, die eine Rolle in der Synthese und Funktion der
Flagelle oder der Biofilmbildung spielen.
Abb. 12: Modell zur Funktion des Tat-Sekretionssystems in A. tumefaciens. Der TatBC-Komplex erkennt ModA (Substratbindeprotein, ABC-Transporter) als Substrat, der TatA-Komplex assoziiert und ModA wird über dessen Kanal ins Periplasma transportiert. Neben der Translokation gefalteter Proteine ist das Tat-System über bisher unbekannte Mechanismen in die Assemblierung der Flagelle, die Chemotaxis und die Virulenz involviert. In dieser Arbeit werden noch weitere mögliche Einflüsse auf die Biofilmbildung und/oder die Siderophor-Sekretion diskutiert. IM, innere Membran; ÄM, äußere Membran.
Da in der PC-defizienten A. tumefaciens Mutante die Expression der tatABC-Gene
nicht verändert ist, könnte PC einen direkten Einfluss auf die Insertion der TatBC-
und/oder TatA-Komplexe in der inneren Membran haben oder auf die Assoziation
und Dissoziation der Komplexe. Dadurch wäre das Tat-Sekretionssystem nicht
Diskussion
75
funktionsfähig und hätte eine fehlerhafte Flagellen-Assemblierung und eine
veränderte Biofilmbildung zur Folge. Diese Hypothese kann noch durch zwei weitere
Punkte unterstützt werden. Durch Proteomanalysen wurde gezeigt, dass das für
A. tumefaciens postulierte Tat-Substrat ModA (Substratbindeprotein, ABC-
Transporter) in der PC-defizienten Mutante in geringeren Mengen im Vergleich zum
Wildtyp unter den getesteten Bedingungen vorliegt. Außerdem ist die Siderophor-
Sekretion in der PC-defizienten Mutante im Vergleich zum Wildtyp reduziert (Daten
nicht gezeigt, Sonja Klüsener). Eine Beteiligung des Tat-Systems an der Siderophor-
Sekretion wurde für P. aeruginosa bereits nachgewiesen (Ochsner et al., 2002), so
dass auch in A. tumefaciens ein Einfluss des Tat-Systems auf die Siderophor-
Sekretion nicht ausgeschlossen werden kann.
PC könnte als Hauptkomponente in den Membranen von A. tumefaciens indirekt
wichtig für die Assemblierung großer Membran-durchspannender Proteinkomplexe
sein und/oder spezifische Funktionen als Chaperon für ausgewählte Proteine
erfüllen.
4. Virulenzdefekt in Abwesenheit von Phosphatidylcholin in A. tumefaciens
Der bedeutendste Phänotyp der PC-defizienten Mutante ist der Virulenzdefekt
(Wessel et al., 2006). Dieser lässt sich durch die fehlende Expression der für die
Virulenz essentiellen vir-Gene erklären, was in dieser Arbeit durch Microarray- und
Reportergenanalysen auf RNA- sowie durch Proteomstudien auf Proteinebene
bestätigt wurde.
In anderen PC-Biosynthesemutanten wurde ebenfalls eine beeinträchtigte Bakterien-
Wirt-Interaktion beobachtet. B. japonicum pmtA-Mutanten bilden eine ineffiziente
Symbiose mit der Sojabohne G. max aus (Minder et al., 2001) und die Infektion
durch PC-defiziente Mutanten der pathogenen Organismen B. abortus und
L. pneumophila ist stark verringert (Comerci et al., 2006; Conde-Alvarez et al., 2006;
Conover et al., 2008). Dies deutet darauf hin, dass in den verschiedenen PC-
synthetisierenden Organismen PC wichtig für die Interaktion mit dem eukaryotischen
Wirt ist, jedoch sind die Ursachen für den jeweiligen Defekt unterschiedlich.
Interessanterweise weist keine dieser PC-Biosynthesemutanten im Vergleich zu der
A. tumefaciens ΔpmtA Δpcs-Mutante einen vergleichsweise drastischen und
Diskussion
76
spezifischen Defekt in der Virulenz auf, der sich bis auf die Kontrolle der
Genexpression eines bestimmten Regulons zurückführen lässt.
Der Grund für die fehlende Induktion der vir-Genexpression in der PC-defizienten
A. tumefaciens-Mutante liegt vermutlich in der Virulenzkaskade stromaufwärts und
somit bei dem Zwei-Komponentensystem VirA/G (Abb. 6). Interessanterweise
werden beide Gene basal in der Mutante exprimiert, jedoch wird die virA-
Genexpression unter Virulenzbedingungen nicht induziert (Tab. 4). Während
Reportergenstudien mit einer virG-lacZ Fusion keinen Unterschied zwischen den
Stämmen und Kultivierungsbedingungen aufzeigen konnten, war virG in den
Microarrayanalysen im Vergleich zum Wildtyp auch unter nicht-induzierten
Bedingungen um einen Faktor von 89 reduziert (Tab. 4).
Tab. 4: Vergleichende Expressionsanalysen. Zusammenfassung der Microarray- und Reportergenstudien für die Zwei-Komponentensysteme VirA/G und ChvG/I in A. tumefaciens Wildtyp (WT) und PC-defizienter Mutante (ΔΔ) unter nicht-induzierten (-AS) und Virulenz-induzierten (+AS) Bedingungen.
Microarray: Fold Changea β-Galaktosidaseaktivität [MU]b
+AS vs. -AS ∆∆ vs. WT WT ∆∆
WT ∆∆ -AS +AS -AS +AS -AS +AS
virA 16 - - -33 1058 1942 1072 1178
virG 2 - -89 -198 269 277 338 340
chvG/I - - - - 195 146 185 172 a Die Genexpression wurde nach einem Vergleich der verschiedenen Stämme und Kultivierungs-bedingungen als unterschiedlich definiert, wenn ein Fold Change von >2 vorlag; -, Genexpression unterscheidet sich nicht. b Angegeben ist die β-Galaktosidaseaktivität in Miller Units [MU] nach Vermessung der plasmid-kodierten transkriptionellen (virA/G) bzw. translationalen (chvG/I) lacZ-Fusionen.
Die Standardlaborbedingungen für die Virulenzinduktion von A. tumefaciens-Zellen
sind die Zugabe von Acetosyringon als phenolischer Induktor und ein saures
Minimalmedium (Rogowsky et al., 1987; Stachel et al., 1986). Die Azidität ist eine
Voraussetzung für die erfolgreiche Virulenzinduktion unabhängig von weiteren
Faktoren, da bei neutralem pH auch in Anwesenheit von Acetosyringon die
Expression der vir-Gene nicht induziert wird (Stachel et al., 1986). Die Azidität wird
zum einen an der cytoplasmatisch lokalisierten Linkerdomäne von VirA
wahrgenommen und zum anderen VirA-unabhängig an dem P1-Promotor des
Antwortregulators VirG, vermittelt über das Zwei-Komponentensystem ChvG/I (Li et
Diskussion
77
al., 2002; Mantis & Winans, 1993; Winans, 1990). Die stark verminderte virG-
Expression in der PC-defizienten Mutante unter nicht-induzierten und Virulenz-
induzierten Bedingungen könnte auf ein nicht funktionales ChvG/I beruhen. Unter
diesen Umständen wäre keine adäquate VirG-Menge vorhanden, um von VirA
aktiviert zu werden, um letztendlich die vir-Genexpression zu induzieren. Jedoch
zeigen sowohl Reportergenstudien als auch Microarrayanalysen, dass chvG/I in
Abwesenheit von PC vergleichbar zum Wildtyp exprimiert wird (Tab. 4).
Die Sensorkinase VirA ist in der inneren Membran lokalisiert und nimmt die Signale
an verschiedenen Domänen wahr. Monosaccharide werden über die
periplasmatische Domäne erkannt und phenolische Verbindungen sowie die Azidität
an der cytoplasmatisch lokalisierten Linkerdomäne (Chang & Winans, 1992). Durch
die stark veränderte Membranzusammensetzung ist es möglich, dass der Induktor
Acetosyringon nicht mehr über die innere Membran in die Zelle gelangen kann, um
dort die Autophosphorylierung von VirA zu induzieren. Durch Reportergenstudien mit
einer virA-lacZ Fusion wurde gezeigt, dass im Wildtyp die virA-Genexpression durch
die alleinige Zugabe des Monosaccharids Glukose induziert werden konnte, nicht
aber in der PC-defizienten Mutante (Daten nicht gezeigt, Sonja Klüsener). Die
fehlende Induktion kann nicht durch die Abwesenheit des Zucker-bindenden Proteins
ChvE erklärt werden, da es unter nicht-induzierten und Virulenz-induzierten
Bedingungen im Vergleich zum Wildtyp sogar leicht verstärkt exprimiert wird
(Microarray; Daten nicht gezeigt, Stephanie Hacker). Obwohl Glukose an der
periplasmatischen Domäne von VirA erkannt wird und nicht die innere Membran
überwinden muss, findet keine Induktion der virA-Expression in Abwesenheit von PC
statt. Dies deutet drauf hin, dass der Virulenzdefekt nicht nur darauf begründet sein
kann, dass Acetosyringon nicht über die innere Membran in die Zelle gelangt.
Setzt man aufgrund der basalen virA/G-Genexpression voraus, dass beide Proteine
in der Zelle vorhanden sind, so muss der Virulenzdefekt durch eine fehlerhafte
Funktionalität des Zwei-Komponentensystems begründet sein. Dabei könnten
verschiedene Ursachen für die fehlerhafte Funktionalität von VirA/G vorliegen
(Abb. 13). Durch die veränderte Membranzusammensetzung wäre es möglich, dass
die Sensorkinase nicht mehr in der inneren Membran lokalisiert ist und somit die
Virulenzsignale nicht wahrnimmt (Abb. 13A).
Diskussion
78
A) Lokalisation: VirA B) Phosphorylierung: VirA
C) Phosphorylierung: VirG D) Bindung an vir-Box
Abb. 13: Mögliche Ursachen für den Virulenzdefekt in Abwesenheit von Phosphatidylcholin. A) Die Sensorkinase VirA ist nicht in der inneren Membran lokalisiert und kann somit die Signale nicht wahrnehmen. B) VirA bildet keinen funktionellen Dimer, so dass die Autophosphorylierung nicht statt finden kann. C) VirA phosphoryliert den Antwortregulator VirG nicht, so dass dieser nicht aktiviert wird. D) VirG bindet nicht an die vir-Boxen, so dass die Transkription der vir-Gene nicht induziert wird. ÄM, äußere Membran; D, Aspartat; H, Histidin; IM, innere Membran; P, Phosphatrest.
VirA muss für die Autophosphorylierung als Homodimer vorliegen. Die Dimerbildung
der Sensorkinase findet an der Membran unabhängig von den Virulenzsignalen statt
Diskussion
79
(Pan et al., 1993) und man kann nicht ausschließen, dass PC direkt oder die richtige
Membranzusammensetzung den Prozess beeinflusst. So könnte in Abwesenheit von
PC die funktionelle Dimerbildung von VirA gestört sein, wodurch die
Autophosphorylierung verhindert werden würde (Abb. 13B). Ebenso ist es möglich,
dass die Virulenzsignale durch VirA erkannt und die Autophosphorylierung induziert
wird, jedoch der Transfer der Phosphatgruppe auf den Antwortregulator VirG nicht
funktioniert. Dadurch würde VirG nicht aktiviert werden und die Expression der vir-
Gene nicht induzieren (Abb. 13C). Der Virulenzdefekt könnte auch durch eine
fehlende Bindung des Transkriptionsaktivators VirG an die vir-Boxen in den
Promotorbereichen der vir-Gene erklärt werden (Abb. 13D).
Es könnte sein, dass bei den möglichen Ursachen für den Virulenzdefekt in
Abwesenheit von PC die PC-Hydrolyse und die dadurch entstehenden Produkte eine
wichtige Rolle als Signalmoleküle spielen. Für viele bakterielle PL ist eine
physiologische Rolle als wichtiger Virulenzfaktor bekannt. Die Enzyme liegen
entweder Membran-gebunden vor oder werden sekretiert, um bei der Infektion die
Membranen des Wirtes zu zerstören und somit das Eindringen des Bakteriums zu
erleichtern (Istivan & Coloe, 2006). In Helicobacter pylori wurden PLA1-, A2- und C-
Aktivitäten nachgewiesen, die in der Degradation der intestinalen Mukosabarriere
beteiligt sind. Yersinia enterocolitica PLA2-Mutanten zeigen eine stark reduzierte
Virulenz im Mausmodell (Nilius & Malfertheiner, 1996; Schmiel et al., 1998). In
Corynebacterium pseudotuberculosis ist die PLD-Aktivität für die Ausbreitung der
Bakterien vom Infektionsort zu den Lymphknoten essentiell und darüber hinaus wird
die pld-Expression in infizierten Makrophagen stark induziert (McKean et al., 2007).
Am Besten sind PL des Typs C in Bakterien charakterisiert, z.B. α-Toxin aus
Clostridium perfringens, zwei PLC’s aus P. aeruginosa, das β-Toxin aus
Staphylococcus aureus und zwei PLC’s aus Listeria monocytogenes (Songer, 1997;
Titball, 1993). Neben ihrer Wirkung als Membran-zerstörende Komponenten
während der Infektion, können die PLC-Enzyme oder die entstehenden sekundären
Metabolite Signaltransduktionskaskaden über die Induktion der Proteinkinase C und
Matrixmetalloproteinasen oder der Inhibierung der NADPH-Oxidase im Wirt
beeinflussen (Firth et al., 1997; Titball, 1993; Traynor et al., 1993). Bisher wurde der
Einfluss der bakteriellen PL während der Infektion nur auf der Seite des Wirtes
untersucht. Es ist jedoch nicht auszuschließen, dass diese auch im Bakterium eine
Rolle in der Virulenzkaskade spielen. Die PL-Enzyme der Säugetiere sind als
Diskussion
80
Effektor-Enzyme in Signaltransduktionswegen bekannt (Exton, 1994). Außerdem
fungiert die durch die PLD-Aktivität freigesetzte Phosphatidsäure als sekundärer
Botenstoff in vielen zellulären Prozessen und ist u.a. in der Rekrutierung von Kinasen
an der Membran beteiligt (Delon et al., 2004). Ebenso könnten PL-Enzyme und/oder
die entstehenden Produkte der PC-Hydrolyse in A. tumefaciens an der
Membranlokalisation und Dimerbildung von VirA beteiligt sein oder die Aktivität von
VirG beeinflussen. Durch erste Blast-Analysen wurden zwei Gene (atu1630 und
atu2486) im Genom von A. tumefaciens identifiziert, die für potentielle PL des Typs D
kodieren könnten. Ob eine PLD-Aktivität tatsächlich vorliegt und ob diese einen
Einfluss auf Signaltransduktionswege in A. tumefaciens haben könnte, muss
experimentell noch überprüft werden.
5. Einfluss von HspL auf das Typ IV-Sekretionssystem in A. tumefaciens
HspL gehört in die Klasse der kleinen Hitzeschockproteine (Hsp), die sich durch eine
konservierte α-kristalline Domäne und eine geringe Molekulargröße auszeichnen
(Balsiger et al., 2004; Caspers et al., 1995). Sie sind weit verbreitet und kommen in
den drei Domänen Bacteria, Archaea und Eukarya vor. Sie fungieren als ATP-
unabhängige Chaperone, die nicht-native Proteine binden, um sie in einem
faltungsfähigen Zustand zu halten. Dadurch sind die Substrate vor Degradation und
Aggregation in der Zelle geschützt. Dabei bilden die kleinen Hsp oft große, oligomere
Komplexe, die in vielen Fällen für die Chaperon-Funktion wichtig sind (Narberhaus,
2002). Die Funktion der kleinen Hsp spiegelt sich in ihrem Expressionsmuster
wieder. Unter normalen Wachstumsbedingungen werden diese kaum exprimiert,
jedoch findet unter Stress eine schnelle und starke Induktion der Expression statt
(Heidelbach et al., 1993; Jobin et al., 1997; Roy & Nakamoto, 1998).
Im Jahr 2006 wurde durch vergleichende Proteomanalysen die Virulenz-abhängige
Induktion der HspL-Synthese in A. tumefaciens entdeckt (Lai et al., 2006). Diese
wurde durch Reportergenstudien und Western-Blot-Analysen bestätigt. Die
Regulation der hspL-Expression, insbesondere unter Virulenzbedingungen, ist
jedoch nicht vollständig geklärt. Die Induktion der hspL-Expression unter
Hitzeschockbedingungen wird durch den alternativen Sigmafaktor σ32 (RpoH)
vermittelt (Balsiger et al., 2004; Rosen et al., 2002). Hingegen ist die hspL-Induktion
Diskussion
81
unter Virulenzbedingungen σ32-unabhängig, da eine Wildtyp-ähnliche hspL-
Expression in einer rpoH-Mutante detektiert wurde (Tsai et al., 2009). Eine
Regulation über das Zwei-Komponentensystem VirA/G kann ebenfalls
ausgeschlossen werden, da die für eine VirG-Bindung essentiellen vir-Boxen im
hspL-Promotorbereich fehlen. Stattdessen wird die Virulenz-induzierte hspL-
Transkription und die drastische Proteinakkumulation spezifisch in virB-Mutanten
reprimiert, so dass die HspL-Expression unter Virulenzbedingungen von den VirB-
Komponenten des TIVSS abhängig ist.
Die Regulation bakterieller kleiner Hsp ist variabel und kann sich grundlegend in den
Mechanismen voneinander unterscheiden. Beispielsweise wird die Transkription von
L. pneumophila gspA durch σ70- und σ32-abhängige Promotoren reguliert. Die
Induktion der gspA-Expression unter Stressbedingungen und während der
intrazellulären Infektion wird dabei durch den σ32-abhängigen Promotor vermittelt
(Abu Kwaik & Engleberg, 1994). Die Expression von hsp18 aus
Clostridium acetobutylicum wird hingegen durch einen Repressor reguliert, der unter
normalen Wachstumsbedingungen die Transkription blockiert (Bahl et al., 1995).
Neben der positiven und/oder negativen Transkriptionskontrolle sind ebenfalls
posttranskriptionelle Regulationsmechanismen beschrieben. In
Synechococcus vulcanus wird die HspA-Menge über die Stabilität der mRNA
kontrolliert (Roy & Nakamoto, 1998). In B. japonicum wird die Translation der hspA-
mRNA und einigen weiteren Vertretern der Hitzeschockantwort über sog. ROSE
(repression of heat shock gene expression)-Elemente reguliert, deren Funktion durch
die Sekundärstruktur des 5’-untranslatierten Bereichs der mRNA bei verschiedenen
Temperaturen bestimmt wird (Nocker et al., 2001). Die Regulation über ROSE-
Elemente wurde ebenfalls für hspAT1 und hspAT2 in A. tumefaciens nachgewiesen,
jedoch nicht für hspL und hspC (Balsiger et al., 2004).
In A. tumefaciens wurde neben dem σ32-abhängigen hspL-Promotor, kein weiterer
Promotor für die induzierte Expression unter Virulenzbedingungen identifiziert, da
mittels Primer-Extension-Experimenten keine alternativen Transkriptionsstartpunkte
detektiert wurden (Daten nicht gezeigt, Sonja Klüsener). Durch RT-PCR-Experimente
sowie Reportergenstudien mit transkriptionellen und translationalen Fusionen wurde
eine Virulenz-abhängige hspL-Expression mit einem Induktionsfaktor von 1,5-2
gezeigt. Im Gegensatz dazu steigt die HspL-Proteinmenge um einen Faktor von 50
an, so dass ein posttranslationaler Regulationsmechanismus vorzuliegen scheint.
Diskussion
82
Die Regulation der hspL-Expression ist von Komponenten des TIVSS abhängig. In
virB6-, virB8-, und virB11-Deletionsmutanten lässt sich eine stark reduzierte HspL-
Menge unter Virulenzbedingungen nachweisen. Obwohl HspL wiederum nicht die
Transkription der virB-Gene beeinflusst, ist die VirB-Proteinmenge in einer hspL-
Mutante deutlich reduziert. Diese wechselseitige Abhängigkeit der HspL- und VirB-
Proteine lässt auf das folgende Modell zur HspL-Expression unter
Virulenzbedingungen schließen (Abb. 14):
Unter nicht-induzierten Bedingungen findet nur eine basale Transkription und
Translation statt, so dass eine geringe HspL-Menge in der Zelle vorliegt. Diese
wenigen HspL-Proteine werden effizient durch zelluläre Proteasen abgebaut
(Abb. 14A). Unter Virulenz-induzierten Bedingungen steigert sich die Transkription
und Translation um einen Faktor von 2 durch einen bisher unbekannten
Regulationsmechanismus. Dadurch akkumuliert HspL und bildet die für kleine Hsp
typischen Oligomere. Gleichzeitig werden die VirB-Proteine, abhängig von dem Zwei-
Komponentensystem VirA/G, in hoher Anzahl synthetisiert. Die HspL-Oligomere
interagieren mit den VirB-Proteinen, so dass sich ein stabiler Komplex aus HspL und
VirB bildet, der vor Degradation geschützt ist (Abb. 14B).
A) nicht-induziert B) Virulenz-induziert
Abb. 14: Modell zur Regulation der HspL-Expression. Unter Hitzeschock wird die HspL-Expression über den σ 32-abhängigen Promotor induziert. Die Virulenz-abhängige Induktion wird hauptsächlich posttranslational kontrolliert. A) Unter nicht-induzierten Bedingungen wird die basal gebildete HspL-Menge schnell wieder degradiert. B) Unter Virulenz-induzierten Bedingungen akkumulieren mehr HspL-Moleküle und die VirB-Proteine werden VirA/G-abhängig synthetisiert. Die HspL-Oligomere interagieren mit VirB und bilden somit einen stabilen Komplex, der nicht abgebaut werden kann.
Diskussion
83
Das Vorkommen vier verschiedener, kleiner Hsp in A. tumefaciens könnte im Laufe
der Evolution dazu geführt haben, dass sich einzelne hsp-Gene funktionell weiter
entwickelt haben, um eine spezifische Rolle in der Zelle zu übernehmen. In B. subtilis
beispielsweise ist das kleine Hsp CotM spezifisch in die Sporulation involviert, indem
es bei der Sporenwandsynthese mitwirkt (Henriques et al., 1997). In A. tumefaciens
scheint HspL zum einen durch die σ32-abhängige Expression unter Hitzeschock eine
allgemeine Funktion in dem Multi-Chaperon-Netzwerk der Zelle zu übernehmen.
Hierbei halten Hsp-Oligomere unspezifisch Proteine in einem faltungsfähigen
Zustand, damit diese durch ATP-abhängige Chaperone (z.B. DnaK) in den nativen
Zustand zurückgefaltet werden können (Narberhaus, 2002) (Abb. 15). Vor kurzem
wurde für HspL in vitro eine eindeutige Chaperonaktivität und Oligomerisierung
nachgewiesen (Tsai et al., 2010). Zum anderen besitzt HspL eine spezifische
physiologische Bedeutung im Infektionsprozess, indem es die Assemblierung und
Stabilität des für die Virulenz essentiellen TIVSS positiv beeinflusst (Abb. 15). Diese
spezifische Funktion von HspL wird durch folgende Punkte belegt: i) Nur die hspL-
Expression wird unter Virulenzbedingungen induziert und nicht die der Anderen drei
hsp-Gene. ii) Lediglich die Expression der VirB-Proteine, nicht aber die weiterer Vir-
Proteine, ist von HspL abhängig. iii) Der konjugale Transfer des IncQ-Plasmids
RSF1010 über das VirB/D4 TIVSS ist um 30 % in der hspL-Mutante im Vergleich
zum Wildtyp reduziert. Im Gegensatz dazu ist der konjugale Transfer des Ti-Plasmid-
Derivats pJK270 über ein weiteres TIVSS, welches durch das trb-Operon kodiert
wird, in der hspL-Mutante unverändert. iiii) Die Tumorbildung ist nach Infektion mit
der hspL-Mutante um 20-25 % im Vergleich zum Wildtyp reduziert.
Vermutlich sind die VirB6-, VirB8- und VirB11-Proteine direkte Substrate von HspL,
da sie in der hspL-Mutante im Vergleich zum Wildtyp in reduzierter Menge vorliegen.
Unter Virulenz-induzierten Bedingungen werden die VirB-Proteine schnell und in
hoher Anzahl produziert. Um eine Aggregation der VirB-Proteine zu vermeiden,
binden die HspL-Oligomere die VirB-Proteine, um dadurch eine geordnete und
schnelle Assemblierung des TIVSS zu gewährleisten (Abb. 15), bevor die ersten
Abwehrmechanismen der Pflanze einsetzen.
In weiteren Bakterien konnten kleine Hsp ebenfalls in Zusammenhang mit den
jeweiligen Infektionsprozessen gebracht werden. In Mycobacterium tuberculosis
akkumuliert das kleine Hsp Acr1 in der stationären Wachstumsphase, während
Hypoxiebedingungen und der Infektion von Makrophagen (Yuan et al., 1996; Yuan et
Diskussion
84
al., 1998). Dabei wird für Acr1 eine wichtige Rolle als Modulator der Immunantwort
des Wirtes diskutiert (Stewart et al., 2006). Des Weiteren wird die Expression des
18 kDa großen Hsp aus Mycobacterium leprae und des Gsp-Proteins aus
L. pneumophila spezifisch während der intrazellulären Infektion induziert (Abu Kwaik
& Engleberg, 1994; Dellagostin et al., 1995). Bisher ist für keines der beschriebenen
Hsp weder ein Substrat bekannt noch die physiologische Bedeutung während der
Infektion. Damit ist die in dieser Arbeit aufgeklärte Verbindung von HspL mit der
Assemblierung des funktionsfähigen TIVSS während der Infektion von
A. tumefaciens der erste direkte Hinweis für die physiologische Bedeutung kleiner
Hsp in der Virulenz.
Abb. 15: Postulierte Funktionen von HspL in A. tumefaciens. HspL fungiert in der Zelle als Chaperon, um Proteine in einem faltungsfähigen Zustand zu halten. Unter Hitzeschock bindet HspL dabei unspezifisch nicht-native Proteine, die letztendlich durch ATP-abhängige Chaperone wie DnaK in den nativen Zustand überführt werden (linke Seite). Unter Virulenz-induzierten Bedingungen bindet HspL spezifisch VirB-Proteine, um eine geordnete und schnelle Assemblierung des TIVSS zu gewährleisten (rechte Seite).
Zusammenfassung
85
F Zusammenfassung
Die Zusammensetzung bakterieller Membranen zeugt von einer hohen Diversität und
Gewichtung der Lipide zwischen den verschiedenen Arten. Phosphatidylcholin (PC),
welches lange Zeit als typisch eukaryotisches Membranlipid galt, wird für etwa 10 %
aller Bakterien als Membrankomponente postuliert.
In dem phytopathogenen Bakterium Agrobacterium tumefaciens ist PC mit ~23 % ein
Hauptphospholipid und kommt sowohl in der inneren als auch in der äußeren
Membran vor. PC wird in diesem Organismus zum einen durch eine dreifache
Methylierung des Vorläufermoleküls Phosphatidylethanolamin mittels einer
Phospholipid N-Methyltransferase (PmtA) synthetisiert und zum anderen durch
direkte Kondensation von Cholin mit CDP-Diacylglycerin durch die PC-Synthase
(Pcs). Die entsprechenden PC-Biosynthesegene sind auf dem zirkulären
Chromosom lokalisiert und werden von eigenen Promotoren transkribiert. Die pmtA-
und pcs-Genexpression erfolgt konstitutiv und wird nicht durch die Produkte (MMPE,
DMPE und PC) oder die Verfügbarkeit des für den PC-Synthaseweg essentiellen
Cholins beeinflusst. Beide PC-Biosyntheseenzyme sind nach heterologer Expression
in Escherichia coli aktiv und produzieren signifikante PC-Mengen.
Ein großer Anteil der PC-synthetisierenden Bakterien geht symbiontische oder
pathogene Interaktionen mit einem eukaryotischen Wirt ein. PC spielt eine wichtige
Rolle in der Bakterien-Wirt-Interaktion. Eine PC-defiziente A. tumefaciens-Mutante ist
nicht mehr in der Lage eine Tumorbildung in Pflanzen zu induzieren. Dieser
Virulenzdefekt beruht auf der fehlenden Bildung des Typ IV-Sekretionssystems
(TIVSS), einem hochmolekularen Membrankomplex, welcher die agrobakterielle
T-DNA in die Pflanze transportiert. Proteom-, Transkriptom- und Reportergenstudien
belegen, dass die Expression der für die Virulenz und Ausbildung des TIVSS
benötigten vir-Gene in Abwesenheit von PC nicht induziert wird. Außerdem zeigen
vergleichende Transkriptom- und Proteomanalysen von Wildtyp und PC-defizienter
Mutante, dass diverse Membran-assoziierte Prozesse in der Mutante verändert sind,
um weiterhin ein normales Wachstum in Abwesenheit von PC zu gewährleisten. Eine
phänotypische Charakterisierung der Mutante nach Anzucht in Minimalmedium weist
auf eine veränderte Biofilmbildung und eine stark reduzierte Motilität basierend auf
einer geringeren Akkumulation der Flagellenproteine FlaA und FlaB hin. Die
Zusammenfassung
86
Ergebnisse belegen, dass PC als Bestandteil der Membranen für A. tumefaciens von
entscheidender Bedeutung ist, insbesondere für den Infektionsprozess.
Ein weiterer wichtiger Faktor für die Virulenz in A. tumefaciens ist das kleine
Hitzeschockprotein HspL. Es akkumuliert unter Virulenz-induzierten Bedingungen
abhängig von Komponenten des TIVSS und scheint eine Rolle in dessen
Assemblierung zu spielen. Im Vergleich zum Wildtyp ist in einer hspL-Mutante der
durch das TIVSS vermittelte konjugale Transfer um 30 % niedriger und die Infektion
von Kartoffelscheiben um 20-25 % reduziert, wobei die Membranlipid-
zusammensetzung unverändert ist.
Mit PC und HspL werden in dieser Arbeit zwei chromosomal kodierte Faktoren
beschrieben, die für die erfolgreiche Infektion der Pflanze durch A. tumefaciens von
zentraler Bedeutung sind.
Summary
87
G Summary
Bacterial membranes are highly diverse and their lipid composition varies between
different species. Phosphatidylcholine (PC) is the most abundant phospholipid in
eukaryotic membranes whereas only 10 % of all bacteria are predicted to synthesize
PC.
The phytopathogenic bacterium Agrobacterium tumefaciens produces PC as a major
membrane lipid. The amount of PC reaches approx. 23 % of total membrane lipids
and it is distributed in both the inner and outer membranes. A. tumefaciens
synthesizes PC either by three consecutive methylation steps of
phosphatidylethanolamine catalyzed by one phospholipid N-methyltransferase
(PmtA) or by direct condensation of choline and CDP-diacylglycerol via the
PC synthase (Pcs). The responsible PC-biosynthesis genes pmtA and pcs are
located on the circular chromosome and transcribed from their own constitutive
promoter. The genes are not regulated by the products MMPE, DMPE and PC or the
availability of choline, an essential precursor of the PC synthase pathway. Upon
recombinant expression of pmtA and pcs, Escherichia coli produces significant
amounts of PC indicating that the individual proteins carry out the annotated enzyme
functions.
Many of the bacteria with PC-containing membranes interact with eukaryotic hosts.
PC plays an important role in the successful establishment of bacteria-host
interactions. A PC-deficient A. tumefaciens mutant is incapable of tumor formation.
The virulence defect is based on the absence of the membrane-spanning type IV
secretion system (TIVSS) which transfers the agrobacterial T-DNA into the plant cell.
Proteome, transcriptome and reportergene analyses of a PC-deficient mutant
confirmed the missing virulence-induced expression of all vir-genes encoding the
essential virulence factors and the TIVSS. Furthermore, the comparative proteomic
and transcriptomic studies with wild-type and PC-deficient mutant strains showed
that diverse processes especially those regarding membrane functions are changed
in the mutant to ensure normal growth in the absence of PC. Additionally, loss of PC
results in altered biofilm formation and reduced motility when cells are grown in
minimal medium. The motility defect is due to reduced levels of the flagellar proteins
Summary
88
FlaA and FlaB. This work documents the critical importance of PC in A. tumefaciens
membranes and its role in diverse cellular processes particularly in virulence.
In A. tumefaciens the small heat shock protein HspL is another critical virulence
factor. Under virulence-induced conditions the HspL protein accumulates in
dependence of components of the TIVSS and presumably supports its correct
assembly. In the hspL mutant the conjugal transfer efficiency via the TIVSS is
reduced by 30 % and tumorigenesis efficiency is 20-25 % lower as compared to the
wild type whereas the lipid composition is wild-type like.
This work describes the two chromosomally encoded factors PC and HspL as
important determinants for successful plant infection by A. tumefaciens.
Referenzen
89
H Referenzen
Abu Kwaik Y. and N.C. Engleberg (1994). Cloning and molecular characterization of a Legionella pneumophila gene induced by intracellular infection and by various in vitro stress conditions. Mol Microbiol 13(2):243-51.
Aktas M. and F. Narberhaus (2009). In vitro characterization of the enzyme
properties of the phospholipid N-methyltransferase PmtA from Agrobacterium tumefaciens. J Bacteriol 191(7):2033-41.
Ames G.F. (1968). Lipids of Salmonella typhimurium and Escherichia coli: structure
and metabolism. J Bacteriol 95(3):833-43. Anand A., S.R. Uppalapati, C.M. Ryu, S.N. Allen, L. Kang, Y. Tang and K.S. Mysore
(2008). Salicylic acid and systemic acquired resistance play a role in attenuating crown gall disease caused by Agrobacterium tumefaciens. Plant Physiol 146(2):703-15.
Arondel V., C. Benning and C.R. Somerville (1993). Isolation and functional
expression in Escherichia coli of a gene encoding phosphatidylethanolamine methyltransferase (EC 2.1.1.17) from Rhodobacter sphaeroides. J Biol Chem 268(21):16002-8.
Bahl H., H. Müller, S. Behrens, H. Joseph and F. Narberhaus (1995). Expression of
heat shock genes in Clostridium acetobutylicum. FEMS Microbiol Rev 17(3):341-8.
Balsiger S., C. Ragaz, C. Baron and F. Narberhaus (2004). Replicon-specific
regulation of small heat shock genes in Agrobacterium tumefaciens. J Bacteriol 186(20):6824-9.
Baron C., O.C. D and E. Lanka (2002). Bacterial secrets of secretion:
EuroConference on the biology of type IV secretion processes. Mol Microbiol 43(5):1359-65.
Bauldry S.A., K.L. Elsey and D.A. Bass (1992). Activation of NADPH oxidase and
phospholipase D in permeabilized human neutrophils. Correlation between oxidase activation and phosphatidic acid production. J Biol Chem 267(35):25141-52.
Beijersbergen A., S.J. Smith and P.J. Hooykaas (1994). Localization and topology of
VirB proteins of Agrobacterium tumefaciens. Plasmid 32(2):212-8. Billah M.M. and J.C. Anthes (1990). The regulation and cellular functions of
phosphatidylcholine hydrolysis. Biochem J 269(2):281-91. Blusztajn J.K., M. Liscovitch and U.I. Richardson (1987). Synthesis of acetylcholine
from choline derived from phosphatidylcholine in a human neuronal cell line. Proc Natl Acad Sci U S A 84(15):5474-7.
Referenzen
90
Bogdanov M., P.N. Heacock and W. Dowhan (2002). A polytopic membrane protein
displays a reversible topology dependent on membrane lipid composition. Embo J 21(9):2107-16.
Boumann H.A., J. Gubbens, M.C. Koorengevel, C.S. Oh, C.E. Martin, A.J. Heck, J.
Patton-Vogt, S.A. Henry, B. de Kruijff and A.I. de Kroon (2006). Depletion of phosphatidylcholine in yeast induces shortening and increased saturation of the lipid acyl chains: evidence for regulation of intrinsic membrane curvature in a eukaryote. Mol Biol Cell 17(2):1006-17.
Bremer J. and D.M. Greenberg (1959). Mono- and dimethylethanolamine isolated
from rat-liver phospholipids. Biochim Biophys Acta 35:287-8. Brencic A., Q. Xia and S.C. Winans (2004). VirA of Agrobacterium tumefaciens is an
intradimer transphosphorylase and can actively block vir gene expression in the absence of phenolic signals. Mol Microbiol 52(5):1349-62.
Cangelosi G.A., R.G. Ankenbauer and E.W. Nester (1990). Sugars induce the
Agrobacterium virulence genes through a periplasmic binding protein and a transmembrane signal protein. Proc Natl Acad Sci U S A 87(17):6708-12.
Caspers G.J., J.A. Leunissen and W.W. de Jong (1995). The expanding small heat-
shock protein family, and structure predictions of the conserved "α-crystallin domain". J Mol Evol 40(3):238-48.
Chang C.H. and S.C. Winans (1992). Functional roles assigned to the periplasmic,
linker, and receiver domains of the Agrobacterium tumefaciens VirA protein. J Bacteriol 174(21):7033-9.
Charles T.C. and E.W. Nester (1993). A chromosomally encoded two-component
sensory transduction system is required for virulence of Agrobacterium tumefaciens. J Bacteriol 175(20):6614-25.
Chesnokova O., J.B. Coutinho, I.H. Khan, M.S. Mikhail and C.I. Kado (1997).
Characterization of flagella genes of Agrobacterium tumefaciens, and the effect of a bald strain on virulence. Mol Microbiol 23(3):579-90.
Chou A.Y., J. Archdeacon and C.I. Kado (1998). Agrobacterium transcriptional
regulator Ros is a prokaryotic zinc finger protein that regulates the plant oncogene ipt. Proc Natl Acad Sci U S A 95(9):5293-8.
Christie P.J. (1997). Agrobacterium tumefaciens T-complex transport apparatus: a
paradigm for a new family of multifunctional transporters in eubacteria. J Bacteriol 179(10):3085-94.
Christie P.J. and E. Cascales (2005). Structural and dynamic properties of bacterial
type IV secretion systems (review). Mol Membr Biol 22(1-2):51-61. Clark J.D., L.L. Lin, R.W. Kriz, C.S. Ramesha, L.A. Sultzman, A.Y. Lin, N. Milona and
J.L. Knopf (1991). A novel arachidonic acid-selective cytosolic PLA2 contains a
Referenzen
91
Ca2+-dependent translocation domain with homology to PKC and GAP. Cell 65(6):1043-51.
Clark J.D., N. Milona and J.L. Knopf (1990). Purification of a 110-kilodalton cytosolic
phospholipase A2 from the human monocytic cell line U937. Proc Natl Acad Sci U S A 87(19):7708-12.
Close T.J., P.M. Rogowsky, C.I. Kado, S.C. Winans, M.F. Yanofsky and E.W. Nester
(1987). Dual control of Agrobacterium tumefaciens Ti plasmid virulence genes. J Bacteriol 169(11):5113-8.
Comerci D.J., S. Altabe, D. de Mendoza and R.A. Ugalde (2006). Brucella abortus
synthesizes phosphatidylcholine from choline provided by the host. J Bacteriol 188(5):1929-34.
Conde-Alvarez R., M.J. Grilló, S.P. Salcedo, M.J. de Miguel, E. Fugier, J.P. Gorvel, I.
Moriyón and M. Iriarte (2006). Synthesis of phosphatidylcholine, a typical eukaryotic phospholipid, is necessary for full virulence of the intracellular bacterial parasite Brucella abortus. Cell Microbiol 8(8):1322-35.
Conover G.M., F. Martínez-Morales, M.I. Heidtman, Z.Q. Luo, M. Tang, C. Chen, O.
Geiger and R.R. Isberg (2008). Phosphatidylcholine synthesis is required for optimal function of Legionella pneumophila virulence determinants. Cell Microbiol 10(2):514-28.
Cronan J.E. (2003). Bacterial membrane lipids: where do we stand? Annu Rev
Microbiol 57:203-24. Cronan J.E., Jr. and E.P. Gelmann (1975). Physical properties of membrane lipids:
biological relevance and regulation. Bacteriol Rev 39(3):232-56. Crooke E., C.E. Castuma and A. Kornberg (1992). The chromosome origin of
Escherichia coli stabilizes DnaA protein during rejuvenation by phospholipids. J Biol Chem 267(24):16779-82.
De Buck E., D. Hoper, E. Lammertyn, M. Hecker and J. Anne (2008a). Differential 2-
D protein gel electrophoresis analysis of Legionella pneumophila wild type and Tat secretion mutants. Int J Med Microbiol 298(5-6):449-61.
De Buck E., E. Lammertyn and J. Anne (2008b). The importance of the twin-arginine
translocation pathway for bacterial virulence. Trends Microbiol 16(9):442-53. De Buck E., L. Maes, E. Meyen, L. Van Mellaert, N. Geukens, J. Anne and E.
Lammertyn (2005). Legionella pneumophila Philadelphia-1 tatB and tatC affect intracellular replication and biofilm formation. Biochem Biophys Res Commun 331(4):1413-20.
de Iannino N.I. and R.A. Ugalde (1989). Biochemical characterization of avirulent
Agrobacterium tumefaciens chvA mutants: synthesis and excretion of β-(1-2)glucan. J Bacteriol 171(5):2842-9.
Referenzen
92
de Kroon A.I. (2007). Metabolism of phosphatidylcholine and its implications for lipid acyl chain composition in Saccharomyces cerevisiae. Biochim Biophys Acta 1771(3):343-52.
de Rudder K.E., I.M. López-Lara and O. Geiger (2000). Inactivation of the gene for
phospholipid N-methyltransferase in Sinorhizobium meliloti: phosphatidylcholine is required for normal growth. Mol Microbiol 37(4):763-72.
de Rudder K.E., C. Sohlenkamp and O. Geiger (1999). Plant-exuded choline is used
for rhizobial membrane lipid biosynthesis by phosphatidylcholine synthase. J Biol Chem 274(28):20011-6.
de Rudder K.E., J.E. Thomas-Oates and O. Geiger (1997). Rhizobium meliloti
mutants deficient in phospholipid N-methyltransferase still contain phosphatidylcholine. J Bacteriol 179(22):6921-8.
de Vrije T., R.L. de Swart, W. Dowhan, J. Tommassen and B. de Kruijff (1988).
Phosphatidylglycerol is involved in protein translocation across Escherichia coli inner membranes. Nature 334(6178):173-5.
Dellagostin O.A., G. Esposito, L.J. Eales, J.W. Dale and J. McFadden (1995). Activity
of mycobacterial promoters during intracellular and extracellular growth. Microbiology 141(8):1785-92.
Delon C., M. Manifava, E. Wood, D. Thompson, S. Krugmann, S. Pyne and N.T.
Ktistakis (2004). Sphingosine kinase 1 is an intracellular effector of phosphatidic acid. J Biol Chem 279(43):44763-74.
Dewey R.E., R.F. Wilson, W.P. Novitzky and J.H. Goode (1994). The AAPT1 gene of
soybean complements a cholinephosphotransferase-deficient mutant of yeast. Plant Cell 6(10):1495-507.
Diez E. and S. Mong (1990). Purification of a phospholipase A2 from human
monocytic leukemic U937 cells. Calcium-dependent activation and membrane association. J Biol Chem 265(24):14654-61.
Ding Z. and P.J. Christie (2003). Agrobacterium tumefaciens twin-arginine-dependent
translocation is important for virulence, flagellation, and chemotaxis but not type IV secretion. J Bacteriol 185(3):760-71.
Djamei A., A. Pitzschke, H. Nakagami, I. Rajh and H. Hirt (2007). Trojan horse
strategy in Agrobacterium transformation: abusing MAPK defense signaling. Science 318(5849):453-6.
Doerrler W.T. (2006). Lipid trafficking to the outer membrane of Gram-negative
bacteria. Mol Microbiol 60(3):542-52. Douglas C.J., R.J. Staneloni, R.A. Rubin and E.W. Nester (1985). Identification and
genetic analysis of an Agrobacterium tumefaciens chromosomal virulence region. J Bacteriol 161(3):850-60.
Referenzen
93
Dupont L., I. Garcia, M.C. Poggi, G. Alloing, K. Mandon and D. Le Rudulier (2004). The Sinorhizobium meliloti ABC transporter Cho is highly specific for choline and expressed in bacteroids from Medicago sativa nodules. J Bacteriol 186(18):5988-96.
Durieux M.E. and K.R. Lynch (1993). Signalling properties of lysophosphatidic acid.
Trends Pharmacol Sci 14(6):249-54. Exton J.H. (1990). Signaling through phosphatidylcholine breakdown. J Biol Chem
265(1):1-4. Exton J.H. (1994). Phosphatidylcholine breakdown and signal transduction. Biochim
Biophys Acta 1212(1):26-42. Firth J.D., E.E. Putnins, H. Larjava and V.-J. Uitto (1997). Bacterial phospholipase C
upregulates matrix metalloproteinase expression by cultured epithelial cells. Infect Immun 65(12):4931-6.
Garfinkel D.J. and E.W. Nester (1980). Agrobacterium tumefaciens mutants affected
in crown gall tumorigenesis and octopine catabolism. J Bacteriol 144(2):732-43.
Gaynor P.M., T. Gill, S. Toutenhoofd, E.F. Summers, P. McGraw, M.J. Homann, S.A.
Henry and G.M. Carman (1991). Regulation of phosphatidylethanolamine methyltransferase and phospholipid methyltransferase by phospholipid precursors in Saccharomyces cerevisiae. Biochim Biophys Acta 1090(3):326-32.
Geiger O., V. Röhrs, B. Weissenmayer, T.M. Finan and J.E. Thomas-Oates (1999).
The regulator gene phoB mediates phosphate stress-controlled synthesis of the membrane lipid diacylglyceryl-N,N,N-trimethylhomoserine in Rhizobium (Sinorhizobium) meliloti. Mol Microbiol 32(1):63-73.
Gelvin S.B. (2003). Agrobacterium-mediated plant transformation: the biology behind
the "gene-jockeying" tool. Microbiol Mol Biol Rev 67(1):16-37. Goodner B., G. Hinkle, S. Gattung, N. Miller, M. Blanchard, B. Qurollo, B.S.
Goldman, Y. Cao, M. Askenazi, C. Halling, et al. (2001). Genome sequence of the plant pathogen and biotechnology agent Agrobacterium tumefaciens C58. Science 294(5550):2323-8.
Gottesman S. and D. Zipser (1978). Deg phenotype of Escherichia coli lon mutants. J
Bacteriol 133(2):844-51. Ha K.S. and J.H. Exton (1993). Activation of actin polymerization by phosphatidic
acid derived from phosphatidylcholine in IIC9 fibroblasts. J Cell Biol 123(6):1789-96.
Hacker S., J. Gödeke, A. Lindemann, S. Mesa, G. Pessi and F. Narberhaus (2008a).
Global consequences of phosphatidylcholine reduction in Bradyrhizobium japonicum. Mol Genet Genomics 280(1):59-72.
Referenzen
94
Hacker S., C. Sohlenkamp, M. Aktas, O. Geiger and F. Narberhaus (2008b). Multiple
phospholipid N-methyltransferases with distinct substrate specificities are encoded in Bradyrhizobium japonicum. J Bacteriol 190(2):571-80.
Hagio M., Z. Gombos, Z. Varkonyi, K. Masamoto, N. Sato, M. Tsuzuki and H. Wada
(2000). Direct evidence for requirement of phosphatidylglycerol in photosystem II of photosynthesis. Plant Physiol 124(2):795-804.
Hanada T., Y. Kashima, A. Kosugi, Y. Koizumi, F. Yanagida and S. Udaka (2001). A
gene encoding phosphatidylethanolamine N-methyltransferase from Acetobacter aceti and some properties of its disruptant. Biosci Biotechnol Biochem 65(12):2741-8.
Hatch G.M., K. O and P.C. Choy (1989). Regulation of phosphatidylcholine
metabolism in mammalian hearts. Biochem Cell Biol 67(2-3):67-77. Haydock S.F., J.A. Dowson, N. Dhillon, G.A. Roberts, J. Cortes and P.F. Leadlay
(1991). Cloning and sequence analysis of genes involved in erythromycin biosynthesis in Saccharopolyspora erythraea: sequence similarities between EryG and a family of S-adenosylmethionine-dependent methyltransferases. Mol Gen Genet 230(1-2):120-8.
Heidelbach M., H. Skladny and H.U. Schairer (1993). Heat shock and development
induce synthesis of a low-molecular-weight stress-responsive protein in the myxobacterium Stigmatella aurantiaca. J Bacteriol 175(22):7479-82.
Henriques A.O., B.W. Beall and C.P. Moran, Jr. (1997). CotM of Bacillus subtilis, a
member of the α-crystallin family of stress proteins, is induced during development and participates in spore outer coat formation. J Bacteriol 179(6):1887-97.
Hermans M.A., B. Neuss and H. Sahm (1991). Content and composition of
hopanoids in Zymomonas mobilis under various growth conditions. J Bacteriol 173(17):5592-5.
Hindahl M.S. and B.H. Iglewski (1984). Isolation and characterization of the
Legionella pneumophila outer membrane. J Bacteriol 159(1):107-13. Ichihashi N., K. Kurokawa, M. Matsuo, C. Kaito and K. Sekimizu (2003). Inhibitory
effects of basic or neutral phospholipid on acidic phospholipid-mediated dissociation of adenine nucleotide bound to DnaA protein, the initiator of chromosomal DNA replication. J Biol Chem 278(31):28778-86.
Ingram L.O. (1977). Preferential inhibition of phosphatidylethanolamine synthesis in
E. coli by alcohols. Can J Microbiol 23(6):779-89. Istivan T.S. and P.J. Coloe (2006). Phospholipase A in Gram-negative bacteria and
its role in pathogenesis. Microbiology 152(5):1263-74.
Referenzen
95
Jin S.G., R.K. Prusti, T. Roitsch, R.G. Ankenbauer and E.W. Nester (1990). Phosphorylation of the VirG protein of Agrobacterium tumefaciens by the autophosphorylated VirA protein: essential role in biological activity of VirG. J Bacteriol 172(9):4945-50.
Jobin M.P., F. Delmas, D. Garmyn, C. Divies and J. Guzzo (1997). Molecular
characterization of the gene encoding an 18-kilodalton small heat shock protein associated with the membrane of Leuconostoc oenos. Appl Environ Microbiol 63(2):609-14.
Johnson R.C., B.P. Livermore, H.M. Jenkin and L. Eggebraten (1970). Lipids of
Treponema pallidum Kazan 5. Infect Immun 2(5):606-9. Judd P.K., R.B. Kumar and A. Das (2005). Spatial location and requirements for the
assembly of the Agrobacterium tumefaciens type IV secretion apparatus. Proc Natl Acad Sci U S A 102(32):11498-503.
Kado C.I. (2002). Negative transcriptional regulation of virulence and oncogenes of
the Ti plasmid by Ros bearing a conserved C2H2-zinc finger motif. Plasmid 48(3):179-85.
Kaneshiro T. and J.H. Law (1964). Phosphatidylcholine synthesis in Agrobacterium
tumefaciens. I. Purification and properties of a phosphatidylethanolamine N-methyltransferase. J Biol Chem 239:1705-13.
Kanfer J.N. (1980). The base exchange enzymes and phospholipase D of
mammalian tissue. Can J Biochem 58(12):1370-80. Kanipes M.I. and S.A. Henry (1997). The phospholipid methyltransferases in yeast.
Biochim Biophys Acta 1348(1-2):134-41. Kennedy E.P. and S.B. Weiss (1956). The function of cytidine coenzymes in the
biosynthesis of phospholipides. J Biol Chem 222(1):193-214. Kent C. (1990). Regulation of phosphatidylcholine biosynthesis. Prog Lipid Res
29(2):87-105. Kent C., P. Gee, S.Y. Lee, X. Bian and J.C. Fenno (2004). A CDP-choline pathway
for phosphatidylcholine biosynthesis in Treponema denticola. Mol Microbiol 51(2):471-81.
Kodaki T., K. Hosaka, J. Nikawa and S. Yamashita (1991). Identification of the
upstream activation sequences responsible for the expression and regulation of the PEM1 and PEM2 genes encoding the enzymes of the phosphatidylethanolamine methylation pathway in Saccharomyces cerevisiae. J Biochem 109(2):276-87.
Krall L., M. Raschke, M.H. Zenk and C. Baron (2002a). The Tzs protein from
Agrobacterium tumefaciens C58 produces zeatin riboside 5'-phosphate from 4-hydroxy-3-methyl-2-(E)-butenyl diphosphate and AMP. FEBS Lett 527(1-3):315-8.
Referenzen
96
Krall L., U. Wiedemann, G. Unsin, S. Weiss, N. Domke and C. Baron (2002b).
Detergent extraction identifies different VirB protein subassemblies of the type IV secretion machinery in the membranes of Agrobacterium tumefaciens. Proc Natl Acad Sci U S A 99(17):11405-10.
Kunsman J.E. (1973). Characterization of the lipids of six strains of Bacteroides
ruminicola. J Bacteriol 113(3):1121-6. Lai E.M., H.W. Shih, S.R. Wen, M.W. Cheng, H.H. Hwang and S.H. Chiu (2006).
Proteomic analysis of Agrobacterium tumefaciens response to the vir gene inducer acetosyringone. Proteomics 6(14):4130-6.
Lamping E., J. Lückl, F. Paltauf, S.A. Henry and S.D. Kohlwein (1994). Isolation and
characterization of a mutant of Saccharomyces cerevisiae with pleiotropic deficiencies in transcriptional activation and repression. Genetics 137(1):55-65.
Lavander M., S.K. Ericsson, J.E. Bröms and A. Forsberg (2006). The twin arginine
translocation system is essential for virulence of Yersinia pseudotuberculosis. Infect Immun 74(3):1768-76.
LeBlanc M.J., V. Gavino, A. Perea, I.M. Yousef, E. Levy and B. Tuchweber (1998).
The role of dietary choline in the beneficial effects of lecithin on the secretion of biliary lipids in rats. Biochim Biophys Acta 1393(2-3):223-34.
Li L., Y. Jia, Q. Hou, T.C. Charles, E.W. Nester and S.Q. Pan (2002). A global pH
sensor: Agrobacterium sensor protein ChvG regulates acid-inducible genes on its two chromosomes and Ti plasmid. Proc Natl Acad Sci U S A 99(19):12369-74.
Li Z. and D.E. Vance (2008). Phosphatidylcholine and choline homeostasis. J Lipid
Res 49(6):1187-94. Lill R., W. Dowhan and W. Wickner (1990). The ATPase activity of SecA is regulated
by acidic phospholipids, SecY, and the leader and mature domains of precursor proteins. Cell 60(2):271-80.
Livermore B.P. and R.C. Johnson (1974). Lipids of the Spirochaetales: comparison of
the lipids of several members of the genera Spirochaeta, Treponema, and Leptospira. J Bacteriol 120(3):1268-73.
López C.S., H. Heras, S.M. Ruzal, C. Sanchez-Rivas and E.A. Rivas (1998).
Variations of the envelope composition of Bacillus subtilis during growth in hyperosmotic medium. Curr Microbiol 36(1):55-61.
Ma L.S., J.S. Lin and E.M. Lai (2009). An IcmF-family protein, ImpLM, is an integral
inner membrane protein interacting with ImpKL, and its walker a motif is required for type VI secretion system-mediated Hcp secretion in Agrobacterium tumefaciens. J Bacteriol 191(13):4316-29.
Referenzen
97
Mantis N.J. and S.C. Winans (1993). The chromosomal response regulatory gene chvI of Agrobacterium tumefaciens complements an Escherichia coli phoB mutation and is required for virulence. J Bacteriol 175(20):6626-36.
Marr A.G. and J.L. Ingraham (1962). Effect of Temperature on the Composition of
Fatty Acids in Escherichia coli. J Bacteriol 84(6):1260-7. Martínez-Morales F., M. Schobert, I.M. López-Lara and O. Geiger (2003). Pathways
for phosphatidylcholine biosynthesis in bacteria. Microbiology 149(12):3461-71.
Matthysse A.G. (1983). Role of bacterial cellulose fibrils in Agrobacterium
tumefaciens infection. J Bacteriol 154(2):906-15. Matthysse A.G., M. Marry, L. Krall, M. Kaye, B.E. Ramey, C. Fuqua and A.R. White
(2005). The effect of cellulose overproduction on binding and biofilm formation on roots by Agrobacterium tumefaciens. Mol Plant Microbe Interact 18(9):1002-10.
McKean S.C., J.K. Davies and R.J. Moore (2007). Expression of phospholipase D,
the major virulence factor of Corynebacterium pseudotuberculosis, is regulated by multiple environmental factors and plays a role in macrophage death. Microbiology 153(7):2203-11.
McLean B.G., E.A. Greene and P.C. Zambryski (1994). Mutants of Agrobacterium
VirA that activate vir gene expression in the absence of the inducer acetosyringone. J Biol Chem 269(4):2645-51.
McMahon K.E. and P.M. Farrell (1986). Effect of choline deficiency on lung
phospholipid concentrations in the rat. J Nutr 116(6):936-43. Medeot D.B., C. Sohlenkamp, M.S. Dardanelli, O. Geiger, M.G. de Lema and I.M.
López-Lara (2010). Phosphatidylcholine levels of peanut-nodulating Bradyrhizobium sp. SEMIA 6144 affect cell size and motility. FEMS Microbiol Lett 303(2):123-31.
Melchers L.S., T.J. Regensburg-Tuink, R.B. Bourret, N.J. Sedee, R.A. Schilperoort
and P.J. Hooykaas (1989). Membrane topology and functional analysis of the sensory protein VirA of Agrobacterium tumefaciens. Embo J 8(7):1919-25.
Merritt P.M., T. Danhorn and C. Fuqua (2007). Motility and chemotaxis in
Agrobacterium tumefaciens surface attachment and biofilm formation. J Bacteriol 189(22):8005-14.
Minder A.C., K.E. de Rudder, F. Narberhaus, H.M. Fischer, H. Hennecke and O.
Geiger (2001). Phosphatidylcholine levels in Bradyrhizobium japonicum membranes are critical for an efficient symbiosis with the soybean host plant. Mol Microbiol 39(5):1186-98.
Referenzen
98
Montoya A.L., M.D. Chilton, M.P. Gordon, D. Sciaky and E.W. Nester (1977). Octopine and nopaline metabolism in Agrobacterium tumefaciens and crown gall tumor cells: role of plasmid genes. J Bacteriol 129(1):101-7.
Murakami K. and A. Routtenberg (1985). Direct activation of purified protein kinase C
by unsaturated fatty acids (oleate and arachidonate) in the absence of phospholipids and Ca2+. FEBS Lett 192(2):189-93.
Narberhaus F. (2002). α−crystallin-type heat shock proteins: socializing
minichaperones in the context of a multichaperone network. Microbiol Mol Biol Rev 66(1):64-93.
Nilius M. and P. Malfertheiner (1996). Helicobacter pylori enzymes. Aliment
Pharmacol Ther 10 Suppl 1:65-71. Nocker A., T. Hausherr, S. Balsiger, N.P. Krstulovic, H. Hennecke and F. Narberhaus
(2001). A mRNA-based thermosensor controls expression of rhizobial heat shock genes. Nucleic Acids Res 29(23):4800-7.
Norris V. and I. Fishov (2001). Hypothesis: membrane domains and hyperstructures
control bacterial division. Biochimie 83(1):91-7. Oberpichler I., R. Rosen, A. Rasouly, M. Vugman, E.Z. Ron and T. Lamparter (2008).
Light affects motility and infectivity of Agrobacterium tumefaciens. Environ Microbiol 10(8):2020-9.
Ochsner U.A., A. Snyder, A.I. Vasil and M.L. Vasil (2002). Effects of the twin-arginine
translocase on secretion of virulence factors, stress response, and pathogenesis. Proc Natl Acad Sci U S A 99(12):8312-7.
op den Kamp J.A., I. Redai and L.L. van Deenen (1969). Phospholipid composition of
Bacillus subtilis. J Bacteriol 99(1):298-303. Pan S.Q., T. Charles, S. Jin, Z.L. Wu and E.W. Nester (1993). Preformed dimeric
state of the sensor protein VirA is involved in plant-Agrobacterium signal transduction. Proc Natl Acad Sci U S A 90(21):9939-43.
Rajendran L. and K. Simons (2005). Lipid rafts and membrane dynamics. J Cell Sci
118(6):1099-102. Ramey B.E., A.G. Matthysse and C. Fuqua (2004). The FNR-type transcriptional
regulator SinR controls maturation of Agrobacterium tumefaciens biofilms. Mol Microbiol 52(5):1495-511.
Ridgway N.D. and D.E. Vance (1988). Kinetic mechanism of
phosphatidylethanolamine N-methyltransferase. J Biol Chem 263(32):16864-71.
Ridgway N.D., Z. Yao and D.E. Vance (1989). Phosphatidylethanolamine levels and
regulation of phosphatidylethanolamine N-methyltransferase. J Biol Chem 264(2):1203-7.
Referenzen
99
Ridley A.J. and A. Hall (1992). The small GTP-binding protein Rho regulates the
assembly of focal adhesions and actin stress fibers in response to growth factors. Cell 70(3):389-99.
Rigomier D., J.P. Bohin and B. Lubochinsky (1980). Effects of ethanol and methanol
on lipid metabolism in Bacillus subtilis. J Gen Microbiol 121(1):139-49. Robertson G.T., M.E. Kovach, C.A. Allen, T.A. Ficht and R.M. Roop, II (2000). The
Brucella abortus Lon functions as a generalized stress response protease and is required for wild-type virulence in BALB/c mice. Mol Microbiol 35(3):577-88.
Rogowsky P.M., T.J. Close, J.A. Chimera, J.J. Shaw and C.I. Kado (1987).
Regulation of the vir genes of Agrobacterium tumefaciens plasmid pTiC58. J Bacteriol 169(11):5101-12.
Rosen R., K. Büttner, D. Becher, K. Nakahigashi, T. Yura, M. Hecker and E.Z. Ron
(2002). Heat shock proteome of Agrobacterium tumefaciens: evidence for new control systems. J Bacteriol 184(6):1772-8.
Roy S.K. and H. Nakamoto (1998). Cloning, characterization, and transcriptional
analysis of a gene encoding an α-crystallin-related, small heat shock protein from the thermophilic cyanobacterium Synechococcus vulcanus. J Bacteriol 180(15):3997-4001.
Salzberg L.I. and J.D. Helmann (2008). Phenotypic and transcriptomic
characterization of Bacillus subtilis mutants with grossly altered membrane composition. J Bacteriol 190(23):7797-807.
Schmiel D.H., E. Wagar, L. Karamanou, D. Weeks and V.L. Miller (1998).
Phospholipase A of Yersinia enterocolitica contributes to pathogenesis in a mouse model. Infect Immun 66(8):3941-51.
Sekiguchi K., M. Tsukuda, K. Ogita, U. Kikkawa and Y. Nishizuka (1987). Three
distinct forms of rat brain protein kinase C: differential response to unsaturated fatty acids. Biochem Biophys Res Commun 145(2):797-802.
Shi W., M. Bogdanov, W. Dowhan and D.R. Zusman (1993). The pss and psd genes
are required for motility and chemotaxis in Escherichia coli. J Bacteriol 175(23):7711-4.
Shinkai A., L.H. Mei, H. Tokuda and S. Mizushima (1991). The conformation of SecA,
as revealed by its protease sensitivity, is altered upon interaction with ATP, presecretory proteins, everted membrane vesicles, and phospholipids. J Biol Chem 266(9):5827-33.
Shukla S.D. and S.P. Halenda (1991). Phospholipase D in cell signalling and its
relationship to phospholipase C. Life Sci 48(9):851-66.
Referenzen
100
Siddiqui R.A. and J.H. Exton (1992). Oleate stimulation of diacylglycerol formation from phosphatidylcholine through effects on phospholipase D and phosphatidate phosphohydrolase. Eur J Biochem 210(2):601-7.
Sohlenkamp C., K.E. de Rudder, V.V. Röhrs, I.M. López-Lara and O. Geiger (2000).
Cloning and characterization of the gene for phosphatidylcholine synthase. J Biol Chem 275(35):18919-25.
Sohlenkamp C., I.M. López-Lara and O. Geiger (2003). Biosynthesis of
phosphatidylcholine in bacteria. Prog Lipid Res 42(2):115-62. Songer J.G. (1997). Bacterial phospholipases and their role in virulence. Trends
Microbiol 5(4):156-61. Stachel S.E. and E.W. Nester (1986). The genetic and transcriptional organization of
the vir region of the A6 Ti plasmid of Agrobacterium tumefaciens. Embo J 5(7):1445-54.
Stachel S.E., E.W. Nester and P.C. Zambryski (1986). A plant cell factor induces
Agrobacterium tumefaciens vir gene expression. Proc Natl Acad Sci U S A 83(2):379-83.
Stachel S.E. and P.C. Zambryski (1986). VirA and VirG control the plant-induced
activation of the T-DNA transfer process of A. tumefaciens. Cell 46(3):325-33. Stewart J.N., H.N. Rivera, R. Karls, F.D. Quinn, J. Roman and C.A. Rivera-Marrero
(2006). Increased pathology in lungs of mice after infection with an α-crystallin mutant of Mycobacterium tuberculosis: changes in cathepsin proteases and certain cytokines. Microbiology 152(1):233-44.
Su S., B.B. Stephens, G. Alexandre and S.K. Farrand (2006). Lon protease of the α-
proteobacterium Agrobacterium tumefaciens is required for normal growth, cellular morphology and full virulence. Microbiology 152(4):1197-207.
Summers M.L., L.M. Botero, S.C. Busse and T.R. McDermott (2000). The
Sinorhizobium meliloti Lon protease is involved in regulating exopolysaccharide synthesis and is required for nodulation of alfalfa. J Bacteriol 182(9):2551-8.
Tahara Y., T. Yamashita, A. Sogabe and Y. Ogawa (1994). Isolation and
characterization of Zymomonas mobilis mutant defective in phosphatidylethanolamine N-methyltransferase. J Gen Appl Microbiol 40(5):389-96.
Tang Y. and R.I. Hollingsworth (1998). Regulation of lipid synthesis in
Bradyrhizobium japonicum: low oxygen concentrations trigger phosphatidylinositol biosynthesis. Appl Environ Microbiol 64(5):1963-6.
Titball R.W. (1993). Bacterial phospholipases C. Microbiol Rev 57(2):347-66.
Referenzen
101
Tornabene T.G. (1973). Lipid composition of selected strains of Yersinia pestis and Yersinia pseudotuberculosis. Biochim Biophys Acta 306(2):173-85.
Traynor A.E., S.A. Weitzman and L.I. Gordon (1993). Bacterial phosphatidylcholine-
preferring phospholipase C reversibly inhibits the membrane component of the NADPH oxidase in human polymorphonuclear leukocytes: implications for host defense. Cell Immunol 152(2):582-93.
Tsai Y.L., Y.R. Chiang, F. Narberhaus, C. Baron and E.M. Lai (2010). The small
heat-shock protein HspL is a VirB8 chaperone promoting type IV secretion-mediated DNA transfer and virulence in Agrobacterium tumefaciens. J Biol Chem, in minor revision.
Tsai Y.L., M.H. Wang, C. Gao, S. Klüsener, C. Baron, F. Narberhaus and E.M. Lai
(2009). Small heat-shock protein HspL is induced by VirB protein(s) and promotes VirB/D4-mediated DNA transfer in Agrobacterium tumefaciens. Microbiology 155(10):3270-80.
van der Bend R.L., J. de Widt, E.J. van Corven, W.H. Moolenaar and W.J. van
Blitterswijk (1992). The biologically active phospholipid, lysophosphatidic acid, induces phosphatidylcholine breakdown in fibroblasts via activation of phospholipase D. Comparison with the response to endothelin. Biochem J 285 (1):235-40.
Vance D.E. and N.D. Ridgway (1988). The methylation of phosphatidylethanolamine.
Prog Lipid Res 27(1):61-79. Vance J.E. and R. Steenbergen (2005). Metabolism and functions of
phosphatidylserine. Prog Lipid Res 44(4):207-34. Walkey C.J., L.R. Donohue, R. Bronson, L.B. Agellon and D.E. Vance (1997).
Disruption of the murine gene encoding phosphatidylethanolamine N-methyltransferase. Proc Natl Acad Sci U S A 94(24):12880-5.
Weiss S.B., S.W. Smith and E.P. Kennedy (1958). The enzymatic formation of
lecithin from cytidine diphosphate choline and D-1,2-diglyceride. J Biol Chem 231(1):53-64.
Wessel M., S. Klüsener, J. Gödeke, C. Fritz, S. Hacker and F. Narberhaus (2006).
Virulence of Agrobacterium tumefaciens requires phosphatidylcholine in the bacterial membrane. Mol Microbiol 62(3):906-15.
Wilderman P.J., A.I. Vasil, W.E. Martin, R.C. Murphy and M.L. Vasil (2002).
Pseudomonas aeruginosa synthesizes phosphatidylcholine by use of the phosphatidylcholine synthase pathway. J Bacteriol 184(17):4792-9.
Winans S.C. (1990). Transcriptional induction of an Agrobacterium regulatory gene at
tandem promoters by plant-released phenolic compounds, phosphate starvation, and acidic growth media. J Bacteriol 172(5):2433-8.
Referenzen
102
Winans S.C. (1992). Two-way chemical signaling in Agrobacterium-plant interactions. Microbiol Rev 56(1):12-31.
Wood D.W., J.C. Setubal, R. Kaul, D.E. Monks, J.P. Kitajima, V.K. Okura, Y. Zhou, L.
Chen, G.E. Wood, N.F. Almeida, Jr., et al. (2001). The genome of the natural genetic engineer Agrobacterium tumefaciens C58. Science 294(5550):2317-23.
Wu H.Y., P.C. Chung, H.W. Shih, S.R. Wen and E.M. Lai (2008). Secretome analysis
uncovers an Hcp-family protein secreted via a type VI secretion system in Agrobacterium tumefaciens. J Bacteriol 190(8):2841-50.
Yamaji-Hasegawa A. and M. Tsujimoto (2006). Asymmetric distribution of
phospholipids in biomembranes. Biol Pharm Bull 29(8):1547-53. Yang J., I. Sangwan, A. Lindemann, F. Hauser, H. Hennecke, H.M. Fischer and M.R.
O'Brian (2006). Bradyrhizobium japonicum senses iron through the status of haem to regulate iron homeostasis and metabolism. Mol Microbiol 60(2):427-37.
Yao Z.M. and D.E. Vance (1988). The active synthesis of phosphatidylcholine is
required for very low density lipoprotein secretion from rat hepatocytes. J Biol Chem 263(6):2998-3004.
Yao Z.M. and D.E. Vance (1989). Head group specificity in the requirement of
phosphatidylcholine biosynthesis for very low density lipoprotein secretion from cultured hepatocytes. J Biol Chem 264(19):11373-80.
Yuan Q., A. Carle, C. Gao, D. Sivanesan, K.A. Aly, C. Hoppner, L. Krall, N. Domke
and C. Baron (2005). Identification of the VirB4-VirB8-VirB5-VirB2 pilus assembly sequence of type IV secretion systems. J Biol Chem 280(28):26349-59.
Yuan Y., D.D. Crane and C.E. Barry, 3rd (1996). Stationary phase-associated protein
expression in Mycobacterium tuberculosis: function of the mycobacterial α-crystallin homolog. J Bacteriol 178(15):4484-92.
Yuan Y., D.D. Crane, R.M. Simpson, Y.Q. Zhu, M.J. Hickey, D.R. Sherman and C.E.
Barry, 3rd (1998). The 16-kDa α-crystallin (Acr) protein of Mycobacterium tuberculosis is required for growth in macrophages. Proc Natl Acad Sci U S A 95(16):9578-83.
Yuan Z.C., E. Haudecoeur, D. Faure, K.F. Kerr and E.W. Nester (2008a).
Comparative transcriptome analysis of Agrobacterium tumefaciens in response to plant signal salicylic acid, indole-3-acetic acid and γ-amino butyric acid reveals signalling cross-talk and Agrobacterium-plant co-evolution. Cell Microbiol 10(11):2339-54.
Yuan Z.C., P. Liu, P. Saenkham, K. Kerr and E.W. Nester (2008b). Transcriptome
profiling and functional analysis of Agrobacterium tumefaciens reveals a general conserved response to acidic conditions (pH 5.5) and a complex acid-
Referenzen
103
mediated signaling involved in Agrobacterium-plant interactions. J Bacteriol 190(2):494-507.
Zachowski A. (1993). Phospholipids in animal eukaryotic membranes: transverse
asymmetry and movement. Biochem J 294 (1):1-14. Zeisel S.H. (1993). Choline phospholipids: signal transduction and carcinogenesis.
Faseb J 7(6):551-7. Zhang Y.M. and C.O. Rock (2008). Membrane lipid homeostasis in bacteria. Nat Rev
Microbiol 6(3):222-33. Ziemienowicz A. (2001). Odyssey of Agrobacterium T-DNA. Acta Biochim Pol
48(3):623-35. Zupan J.R. and P. Zambryski (1995). Transfer of T-DNA from Agrobacterium to the
plant cell. Plant Physiol 107(4):1041-7.
Publikationen
104
I Publikationen
1. Originalartikel
Wessel M., S. Klüsener, J. Gödeke, C. Fritz, S. Hacker and F. Narberhaus (2006). Virulence of Agrobacterium tumefaciens requires phosphatidylcholine in the bacterial membrane. Mol Microbiol 62(3):906-15.
Klüsener S.*, M. Aktas*, K.M. Thormann, M. Wessel and F. Narberhaus (2009). Expression and physiological relevance of Agrobacterium tumefaciens phosphatidylcholine biosynthesis genes. J Bacteriol 191(1):365-74.
Tsai Y.L., M.H. Wang, C. Gao, S. Klüsener, C. Baron, F. Narberhaus and E.M. Lai (2009). Small heat-shock protein HspL is induced by VirB protein(s) and promotes VirB/D4-mediated DNA transfer in Agrobacterium tumefaciens. Microbiology 155(10):3270-80.
Klüsener S., S. Hacker, Y.L. Tsai, J.E. Bandow, R. Gust, E.M. Lai and F. Narberhaus (2010). Proteomic and transcriptomic characterization of a virulence-deficient phosphatidylcholine-negative Agrobacterium tumefaciens mutant. Mol Genet Genomics 283(6):575-89.
Aktas M., M. Wessel, S. Hacker, S. Klüsener, J. Gleichenhagen and F. Narberhaus (2010). Phosphatidylcholine biosynthesis in plant-pathogenic and symbiotic bacteria. Eur J Cell Biol Rev, submitted. *authors contributed equally
2. Konferenzbeiträge
Hacker S., M. Wessel, M. Aktas, J. Gödeke, S. Klüsener, C. Fritz, C. Sohlenkamp, O. Geiger and F. Narberhaus (2006). Bacterial phosphatidylcholine is required for pathogenic and symbiotic plant-microbe interactions. 2nd FEMS (Federation of European Microbiological Societies) Congress of European Microbiologists, Madrid, Spain, Book of Abstracts p. 71.
Hacker S., M. Wessel, M. Aktas, J. Gödeke, S. Klüsener, C. Fritz and F. Narberhaus (2007). Bacterial phosphatidylcholine is required for pathogenic and symbiotic plant-microbe interactions. International Symposium on Microbial Adaption to Stress and Environment of the Collaborative Research Center 395 (SFB395), Marburg, Germany, Book of Abstracts p. 19.
Publikationen
105
Klüsener S., M. Wessel, M. Aktas, J. Gödeke, C. Fritz and F. Narberhaus (2007). Virulence of Agrobacterium tumefaciens requires phosphatidylcholine in the bacterial membrane. Jahrestagung der Vereinigung für Allgemeine und Angewandte Mikrobiologie (VAAM), Osnabrück, Germany, Book of Abstracts p. 109.
Klüsener S., M. Wessel, S. Hacker, J. Gödeke, C. Fritz and F. Narberhaus (2007). Bacterial phosphatidylcholine is required for pathogenic and symbiotic plant-microbe interactions. XIII. International Congress on Molecular Plant-Microbe Interactions (MPMI), Sorrento, Italy, Book of Abstracts p. 111. Hacker S., M. Aktas, S. Klüsener, M. Wessel, J. Gleichenhagen, C. Fritz and F. Narberhaus (2008). Pathogenic and symbiotic plant-microbe interaction require bacterial phosphatidylcholine. Jahrestagung der Vereinigung für Allgemeine und Angewandte Mikrobiologie (VAAM), Frankfurt, Germany, Book of Abstracts p. 188. Klüsener S., M. Wessel, Y.L. Tsai, E.M. Lai and F. Narberhaus (2008). Virulence of Agrobacterium tumefaciens requires phosphatidylcholine. XII. International Congress of Bacteriology and Applied Microbiology (IUMS), Istanbul, Turkey, Book of Abstracts p. 189.
Klüsener S., M. Aktas, K.M. Thormann, M. Wessel and F. Narberhaus (2009). Expression and physiological relevance of Agrobacterium tumefaciens phosphatidylcholine biosynthesis genes. Jahrestagung der Vereinigung für Allgemeine und Angewandte Mikrobiologie (VAAM), Bochum, Germany, Book of Abstracts p. P111.
Anhang
106
J Anhang
1. Curriculum vitae
Persönliche Angaben Name Sonja Klüsener
Anschrift Dibergstr. 10, 44789 Bochum
Geburtsdatum 09.05.1982
Geburtsort Wuppertal
Familienstand ledig
Staatsangehörigkeit deutsch Schulbildung 08/1988 – 07/1992 Grundschule Siegelberg, Wuppertal 08/1992 – 06/2001 Theodor-Heuss-Gymnasium, Radevormwald
Abschluss Abitur mit Gesamtnote 1,7 Studium 10/2001 – 05/2006 Diplomstudium der Biologie
Ruhr-Universität Bochum Abschluss Diplom mit der Gesamtnote 1,1
Promotion seit 08/2006 Anfertigung der Doktorarbeit
Lehrstuhl für Biologie der Mikroorganismen Ruhr-Universität Bochum
Berufstätigkeit 10/2003 – 11/2003 Studentische Hilfskraft
Lehrstuhl für Zellmorphologie und Molekulare Neurobiologie Ruhr-Universität Bochum
01/2005 – 03/2005 09/2005 – 01/2006
Studentische Hilfskraft Lehrstuhl für Biologie der Mikroorganismen Ruhr-Universität Bochum
seit 08/2006 Wissenschaftliche Mitarbeiterin
Lehrstuhl für Biologie der Mikroorganismen Ruhr-Universität Bochum
Anhang
107
2. Erklärung
Hiermit erkläre ich, dass ich die Arbeit selbständig verfasst und bei keiner anderen
Fakultät eingereicht und dass ich keine anderen als die angegebenen Hilfsmittel
verwendet habe. Es handelt sich bei der heute von mir eingereichten Dissertation um
sechs in Wort und Bild völlig übereinstimmende Exemplare.
Weiterhin erkläre ich, dass digitale Abbildungen nur die originalen Daten enthalten
und in keinem Fall inhaltsverändernde Bildbearbeitung vorgenommen wurde.
Bochum, den 14.04.2010
________________ (Sonja Klüsener)
Anhang
108
3. Eigenanteil an den Publikationen
„Charakterisierung und physiologische Relevanz des Membranlipids Phosphatidylcholin für das pflanzenpathogene Bakterium
Agrobacterium tumefaciens“
Dissertation, Sonja Klüsener, Lehrstuhl für Biologie der Mikroorganismen Referent: Prof. Dr. Franz Narberhaus Korreferent: Prof. Dr. Matthias Rögner Bochum, den 14.04.2010
1. Tsai Y.L., M.H. Wang, C. Gao, S. Klüsener, C. Baron, F. Narberhaus and E.M. Lai (2009). Small heat-shock protein HspL is induced by VirB protein(s) and promotes VirB/D4-mediated DNA transfer in Agrobacterium tumefaciens. Microbiology 155(10):3270-80.
20 % Planung, Durchführung, Auswertung aller Experimente und Verfassen des Manuskripts
2. Klüsener S.*, M. Aktas*, K.M. Thormann, M. Wessel and F. Narberhaus (2009). Expression and physiological relevance of Agrobacterium tumefaciens phosphatidylcholine biosynthesis genes. J Bacteriol 191(1):365-74.
*authors contributed equally
50 % Planung, Durchführung und Auswertung aller Experimente, 80 % Verfassen des Manuskripts
3. Klüsener S., S. Hacker, Y.L. Tsai, J.E. Bandow, R. Gust, E.M. Lai and F. Narberhaus (2010). Proteomic and transcriptomic characterization of a virulence-deficient phosphatidylcholine-negative Agrobacterium tumefaciens mutant. Mol Genet Genomics 283(6):575-89.
80 % Planung, Durchführung, Auswertung aller Experimente und Verfassen des Manuskripts
Recommended