New insights into the role of Bartonella effector proteinsin pathogenesisSabrina Siamer and Christoph Dehio

Available online at www.sciencedirect.com

ScienceDirect

The facultative intracellular bacteria Bartonella spp. share a

common infection strategy to invade and colonize mammals in

a host-specific manner. Following transmission by blood-

sucking arthropods, Bartonella are inoculated in the derma and

then spread, via two sequential enigmatic niches, to the blood

stream where they cause a long-lasting intra-erythrocytic

bacteraemia. The VirB/VirD4 type IV secretion system (VirB/D4

T4SS) is essential for the pathogenicity of most Bartonella

species by injecting an arsenal of effector proteins into host

cells. These bacterial effector proteins share a modular

architecture, comprising domains and/or motifs that confer an

array of functions. Here, we review recent advances in

understanding the function and evolutionary origin of this

fascinating repertoire of host-targeted bacterial effectors.

Addresses

Focal Area Infection Biology, Biozentrum, University of Basel, Basel,

Switzerland

Corresponding author: Dehio, Christoph (christoph.dehio@unibas.ch)

Current Opinion in Microbiology 2015, 23:80–85

This review comes from a themed issue on Host-microbe interac-

tions: bacteria

Edited by David Holden and Dana Philpott

http://dx.doi.org/10.1016/j.mib.2014.11.007

1369-5274/# 2014 Published by Elsevier Ltd.

IntroductionBartonella spp. are facultative intracellular bacteria that

are responsible for long-lasting intra-erythrocytic bacter-

aemia in diverse mammalian reservoir hosts and are

transmitted by blood-sucking arthropods most likely

through inoculation of skin lesions by contaminated

insect feces, or contact with infected animals via scratch-

ing or bites [1]. Following dermal inoculation, Bartonellacolonize two sequential niches, respectively called ‘der-

mal niche’ (describing the dermal stage of infection) and

‘blood-seeding niche’ (formerly known as ‘primary

niche’), considered to include dendritic and endothelial

cells [2��]. Subsequently, Bartonella reach the blood

stream where, restricted to the specific reservoir host,

they invade erythrocytes and persist intracellularly for the

remaining lifetime of the red blood cell [3]. Specific

Current Opinion in Microbiology 2015, 23:80–85

adaptation to the reservoir host causes no or only mild

disease symptoms, while infection of incidental hosts can

be associated with broad spectrum of diseases. The

exception is B. bacilliformis, a human-specific species that

causes life-threatening disease. Most other human infec-

tions are caused by the human-specific species B. quintanaand the zoonotic cat-specific species B. henselae. The

clinical manifestations and treatments of Bartonellainfection have been extensively reviewed elsewhere

and therefore will not be covered in this review [4�,5�].

Different genomic and phylogenetic studies revealed that

the genus Bartonella is divided into four lineages plus

B. tamiae and B. australis, which occupy ancestral posi-

tions [6,7,8��]: lineage 1 is solely composed of the deadly

pathogen B. bacilliformis. Lineage 2 is composed of rumi-

nant-specific species (e.g. deer-specific B. schoenbuchensis)that have acquired the Vbh (VirB-homologous) T4SS that

is closely related to the VirB/D4 T4SS. Lineage 3 is

composed of species infecting diverse mammals (e.g.

fox/dog-specific B. rochalimae and cat-specific B. clarrid-geiae) that have acquired the VirB/D4 T4SS and the most

species-rich lineage 4 contains species (e.g. cat-specific

B. henselae, human-specific B. quintana, or rat-specific

B. tribocorum) that harbor the VirB/D4 and the Trw

T4SS [6,7,8��]. Both T4SS present in the lineage 4 species

were shown to be essential for host interaction at different

stages of the infection cycle [9,10]. The Trw T4SS is

involved in the erythrocyte invasion process by mediating

host-specific adhesion to these cells [10]. The VirB/D4

T4SS contributes to pathogenicity by translocating a

cocktail of evolutionary-related effector proteins, called

Beps (Bartonella effector proteins) into nucleated cells of

both the ‘dermal’ and the ‘blood-seeding’ niches [2��,6,9].

Once inside host cells, these Beps target host components

and modulate cellular processes to the benefit of the

bacteria. Therefore, defining the role of the various Beps

is key to gain a better understanding of the pathogenesis

of Bartonella. This review summarizes advances made in

deciphering the evolution and functional role of the Beps

in Bartonella–host cell interaction.

Acquisition of the VirB/D4 T4SSThe current model indicates that the radiating lineage

3 and 4 have independently acquired the virB/D4 T4SS

locus with at least one effector gene by horizontal gene

transfer [6,7�]. The similarity of the VirB/D4 T4SS with

bacterial conjugation systems suggests that the virB/D4locus was probably acquired via horizontal gene transfer

from a conjugative plasmid. Whole genome shotgun

www.sciencedirect.com

Function of Bartonella effector proteins Siamer and Dehio 81

analysis of representative species of lineages 3 and

4 demonstrated lineage-specific integration of the virB/D4 locus and the presence of lineage-specific repertoires

of fast evolving bep genes that display a high degree of

inter-species variation. Comparative whole genome

analysis further indicated that the virB/D4 and bep gene

clusters represent a major driver of the adaptive radiation

observed in the lineages 3 and 4 which represent a

remarkable example of parallel evolution [6].

Evolution of the BepsBartonella–host interaction exposes the arsenal of Beps to

a strong selective pressure resulting in their rapid evolu-

tion, which could explain their modular architecture. This

modular architecture of the Beps, which is derived from a

single ancestor composed of an N-terminal FIC and C-

terminal BID domain, evolved by extensive rounds of

duplication, diversification and reshuffling of domains

(Figure 1) [6,11]. The BID domain together with a

Figure 1

Independeduplicatio

Ancestral effect

Representative Beps from the lineage 3

Recombination andadaptive muta

FIC motif

FIC B

FIC BID

FIC BID

FIC

FIC

FIC

BID

BIDBID

BID

BID

BID

FIC motif

FIC motif

FIC motif

FIC motif

HPFxxxNG

xPFxxGNx

Parallel evolution of the Beps from the lineages 3 and 4. The Beps from the

the same FIC-BID domain architecture. Subsequently, after lineage-specific

adaptive mutations, the Beps with a derived domain structure emerged inde

www.sciencedirect.com

positively charged C-tail constitute a secretion signal

for the VirB/D4 T4SS, while the FIC, also present in

other bacterial effectors and many more endogenous

proteins found in all domains of life, was shown to

mediate post-translational modifications such as AMPyla-

tion, phosphocholination or phosphorylation [12–15].

Interestingly, previous studies suggest that the BID

domain together with a positively charged C-tail, which

evolved as bi-partitie T4SS signal from relaxases of

alphaproteobacterial conjugation systems, was acquired

by the ancestral Bartonella effector protein to facilitate

injection into target cells [11]. This is supported by the

fact that in B. grahamii the Vbh T4SS, which is closely

related to the VirB/D4 T4SS, is encoded on a plasmid in

addition to a chromosomally encoded copy [16]. The vbhlocus on the plasmid also encodes a toxin called VbhT in

B. schoenbuchensis, which harbors the classical FIC-BID

domain architecture as found conserved in many Beps

[13]. It is thus tempting to speculate that the VbhT toxin

ntn

or VbhT

fixation oftions

Representative Beps from the lineage 4

ID

FIC

FIC FIC

FIC

BID BID

BID

BID BID

BID

BID

BID

BID BID

BID

BID

BID

FIC BID

FIC BID

FIC motif

FIC motif

FIC motifHPFxxGNx

xPFxxGNx

FIC motif

FIC motif

Current Opinion in Microbiology

lineages 3 and 4 evolved possibly from the VbhT toxin, which harbors

amplifications followed by recombination events and fixation of

pendently in both lineages 3 and 4.

Current Opinion in Microbiology 2015, 23:80–85

82 Host-microbe interactions: bacteria

is closely related to the ancestral Bep from which the

arsenal of effectors of both lineages 3 and 4 evolved

independently by successive lineage-specific duplications

and was further shaped by positive selection of diversified

gene copies (Figure 1). The arsenal of Beps can be sorted

into ten phylogenetic clades for the lineage 3, and seven

phylogenetic clades for the lineage 4. The FIC-BID

domain structure, followed by a positively charged C-tail,

displays the most abundant effector protein type. How-

ever, a subset of effectors both in lineages 3 and 4 harbors

other functional modules instead of the FIC domain, like

tandem-repeated tyrosine-phosphorylation motifs (EPIYA

and EPIYA-related motifs) or additional BID domains [11].

These different domain architectures suggest that these

Beps arose from the ancestral domain structure by inde-

pendent recombination events (Figure 1). Thus, the evol-

utionary diversification of the Beps created a wide array of

adaptive functions dedicated to the cellular interaction

within the mammalian hosts. This remodeling of pre-

existing effector proteins as an evolutionary mechanism

to rapidly adapt to the host was described by Stavrinides

et al. as ‘terminal reassortment’ in which effector genes

recombine among themselves or with other genetic

elements to give rise to new chimeric effectors [17].

Interestingly, comparative whole genome analysis of 14

Bartonella genomes revealed the presence of a Bartonellagene transfer agent (BaGTA) that represents the most

conserved key innovation driving the explosive radiations

of both lineages 3 and 4 [7�]. Gene transfer agents (GTAs)

are phage-like elements that generally transfer random

pieces of the bacterial genome rather than their own

DNA. GTA genes are located on the host chromosome,

and GTAs generally transfer DNA from the producing

cell to a recipient cell via a mechanism similar to trans-

duction. Although the function of GTAs remains unclear

regarding their role in evolution, they provide an efficient

mechanism to duplicate and recombine genes [18�]. Bep

gene duplication and recombination by BaGTA may thus

have conferred the high degree of adaptability to diverse

mammalian reservoir hosts that have led to the explosive

radiations of lineages 3 and 4.

Actin remodeling and invasome formationHistorically, B. henselae was found to elicit VirB/D4-de-

pendent phenotypes in vitro using human umbilical vein

endothelial cells (HUVECs), like bacterial internalization

via the invasome structure, inhibition of apoptosis, or

activation of proinflammatory signaling [19]. This inter-

action of Bartonella with endothelial cells in vitro and the

clinical descriptions of B. henselae in association with mani-

festations in the human vasculature, led to the proposition

that vascular endothelial cells represent a part of the

‘blood-seeding niche’, that is colonized initially and from

where bacteria are seeded to the blood stream to invade

erythrocytes [2��]. Studies of the infection of endothelial

cells have shown that the internalization process of

Current Opinion in Microbiology 2015, 23:80–85

Bartonella starts with endocytosis of one or a few bacteria

into a vacuole called Bartonella-containing vacuole (BCV).

This process is then arrested by the activities of several

Beps (i.e. BepC, BepF and BepG in B. henselae), which are

translocated via the VirB/D4 T4SS into the host cell

cytoplasm [20–22]. It was shown that both BepG alone

or a combined action of BepC and BepF impedes BCV

formation via the inhibition of endocytosis. This leads to

accumulation of bacterial aggregates at the cell surface,

which invade the endothelial cells via F-actin-dependent

invasome-mediated internalization [21,22]. Invasome for-

mation mediated by combined action of BepC and BepF

depends on F-actin modulation by Rac1/Scar1/Wave/Arp2/

3, Cdc42/WASP/Arp2/3 and cofilin1. In contrast, BepG-

triggered invasome formation requires only F-actin modu-

lation by Rac1/Scar1/Wave/Arp2/3 and Cdc42/WASP/

Arp2/3, but not cofilin 1 [21,22]. Interestingly, constitu-

tive-active Cdc42 or Rac1 can mimic BepF action in the

BepC/BepF dependent invasome formation pathway,

suggesting a regulatory role of BepF on the small Rho

GTPases. BepF harbors a tyrosine-repeat motif near its

N-terminus and contains three BID domains in the C-

terminal part. Although the N-terminal tyrosine-repeat

phosphorylation motif is phosphorylated upon transloca-

tion via the VirB/D4 T4SS, it is not involved in invasome

formation [23]. Instead, the first two BID domains of BepF,

BID-F1 and BID-F2, are sufficient to promote the for-

mation of these structures [23]. BepG harbors an array of

four BID domains. However, sequence comparisons of

B. henselae BID domains revealed that the BID domains

of BepF and BepG are poorly conserved between each

other suggesting that these effectors may promote inva-

some formation by targeting different host functions [11].

Inhibition of apoptosisBepA mediates protection of endothelial cells from apop-

tosis and may thus contribute to the formation of vaso-

proliferative tumors as found in bacillary angiomatosis, a

typical manifestation of immunocompromised patients

infected with B. henselae or B. quintana [24��,25]. On

the molecular level, the BID domain of BepA was shown

to block endothelial cell apoptosis by direct binding to

host cell adenylyl cyclase, thereby potentiating GaS-de-

pendent cAMP production. The elevation of cAMP levels

and consecutive upregulation of cAMP-stimulated gene

expression then leads to inhibition of apoptosis [24��].While inhibition of apoptosis is exclusively mediated by

the BID domain of BepA, the FIC domain shows auto-

AMPylation and AMPylation of a target from total HeLa

cell extract [26]. The nature of the host target and

physiological consequences of AMPylation remains to

be demonstrated.

BepE: an essential protein for BartonelladisseminationRecently, it was shown that HUVEC cells infected with

the DbepE mutant of B. henselae are deficient in rear end

www.sciencedirect.com

Function of Bartonella effector proteins Siamer and Dehio 83

detachment and undergo cell fragmentation. These phe-

notypes require the action of BepC, which possibly

affects rear detachment during cell migration by modu-

lating the actin cytoskeleton [2��]. Ectopic expression of

BepE from B. henselae or its orthologs in cells is sufficient

to suppress the cell fragmentation phenotype [2��]. The

prominent activity of BepE in restoring cell migration led

the authors to dissect the in vivo function of this effector

protein. They demonstrated using a rat model and differ-

ent routes of infection that BepE is essential to invade

some nucleated cell types in the derma and spread the

infection to the blood stream. As observed for BepA,

BepF and BepG, this BepE-dependent phenotype also

relies on the effector BID domains. The authors assumed

that the migration of Bartonella from the derma to the

blood stream could be achieved intracellularly in den-

dritic cells. This is supported by the fact that a trans-well

migration assay showed an inhibition of dendritic cell

migration upon infection with the B. henselae DbepDEF

Figure 2

Blood-suckingarthropod

Dermis

D

7

8

9 2

1

3

Capillary

Vasculature

Bartonella

Model of the Bartonella infection process in the reservoir hosts. (1) Bartonel

in the feces. (2) When the arthropods suck blood on mammals, it provides

Bartonella-containing insect feces into the derma. (3) Subsequently, Bartone

Ref. [2��]). This ‘dermal niche’ includes most likely dendritic cells, that are c

(formerly known as ‘primary niche’) a process that strictly depends on the e

to colonize endothelial cells, which likely require the action of BepC, BepF,

periodically seeded into the bloodstream where they invade erythrocytes an

replication inside the red blood cell (7), they persist in the ‘intraerythrocytic

competent for transmission by a bloodsucking arthropod (9).

www.sciencedirect.com

mutant, while complementation with BepE restores it

[2��]. These new results on the function of BepE allowed

the authors to extend the concept of the ‘blood-seeding

niche’ (formerly known as ‘primary niche’) to ‘dermal

niche’, which describes the dermal stage of infection

[2��]. Moreover, the authors demonstrated that BepE acts

via its second BID domain on the RhoA signaling path-

way to most likely preserve cells from fragmentation [2��].Indeed, they observed that BepE interferes with the

inhibitory effect of the C3-based Rho inhibitor 1, which

is known to ADP-ribosylate RhoA [27].

Altogether, the experimental studies showed that the

Beps are essential for the primary stage of infection

(Figure 2). It is surprising to see that most Bep-dependent

phenotypes of B. henselae that have been studied so far

rely on the BID domain, which evolved as a secretion

signal and acquired, over the course of evolution, diverse

functions, with distinct phenotypic properties. Until now,

endritic cells

Dermal niche

Blood seeding niche

Migration

BepE

BepF BepG

Invasome formation

4

5

6Inhibition of apoptosis

BepC

BepA lumen

Current Opinion in Microbiology

la replicate first in the midgut of the arthropod vector and are excreted

a local source of irritation followed by scratching and inoculation of

lla colonize the ‘dermal niche’, a term proposed by Okujava et al. (see

onsidered to disseminate Bartonella towards the ‘blood seeding niche’

ffector BepE (4). In the ‘blood-seeding niche’, bacteria are considered

BepG and BepA (5). From this ‘blood seeding niche’, the bacteria are

d reinfect the cells of the ‘blood seeding niche’ (6). After limited

niche’ for the remaining life-span of the red blood cell (8) and are thus

Current Opinion in Microbiology 2015, 23:80–85

84 Host-microbe interactions: bacteria

Table 1

Beps studied and their function in Bartonella pathogenicity

Beps Bartonella species Phenotype Domain implicated Target References

BepA B. henselae Inhibition of apoptosis BID domain Adenylyl cyclase [25,26]

BepC B. henselae F-actin modulation

Trigger invasome formation

Trigger cell fragmentation

Unknown Related to Rac1 and

Cdc42 signaling,

Cofilin1-dependent

[2��,21,22]

BepE B. henselae

B. tribocorum

Preserves ECs from fragmentation

and blocks defects of DCs migration

caused by other Beps

BID domains 1 and 2 RhoA signaling pathway [2��]

BepE B. henselae Unknown Tyrosine-phosphorylation

motif

Csk

SHP2

[28]

BepF B. henselae F-actin modulation

Trigger invasome formation

BID domains 1 and 2 Related Rac1 and Cdc42

signaling, Cofilin1-dependent

[21–23]

BepG B. henselae F-actin modulation

Trigger invasome formation

Multiple BID domains Related to Rac1 and Cdc42

signaling, Cofilin1-independent

[21,22]

Bep2 B. rochalimae Unknown FIC domain AMPylation of Vimentin [29��]

Abbreviations: ECs, endothelial cells; DCs, dendritic cells.

no phenotype in vivo or in vitro has been shown to depend

on the tyrosine phosphorylation motifs or the FIC domain

of any Bep from lineage 4. BepD, BepE and BepF from

the lineage 4 and Bep9 from the lineage 3 were shown to

be tyrosine-phosphorylated in human cells [6,11,28].

More particularly, using a proteomic screen to identify

potential targets of all known tyrosine-phosphorylated

bacterial effectors, BepE was shown to interact with

Csk upon phosphorylation on Tyr37 and with SHP2 upon

phosphorylation with Tyr64. Although this result was

confirmed by a co-immunopurification assay, further work

is required to elucidate the biological relevance of these

tyrosine-phosphorylation and subsequent protein inter-

actions [28]. All the Beps discussed and their implications

in Bartonella pathogenicity are listed in Table 1.

Beps from the lineage 3: opening the blackboxWhile the Bep effectors of lineage 4 and particularly of

B. henselae have been extensively studied, the functions

of the Beps of lineage 3 are still poorly understood. As

described above, these Beps are classified in 10 phyloge-

netic clades and most of them, except the clade 9, harbor

the classical FIC-BID domain organization [6]. Recently,

Pieles et al. showed by an unbiased mass spectrometry-

based approach that Bep2 from B. rochalimae AMPylates

via its FIC domain the intermediate filament protein

vimentin [29��]. Although the contribution to virulence

by this AMPylation remains to be elucidated, this method

paves the way for identifying AMPylation targets for

other FIC domains of Beps from lineages 3 and 4. This

approach can further be adapted to study other post-

translational modifications for which isotopically labeled

substrates are available. As it was shown that FIC domains

can mediate diverse post-translational modifications

[12–15], this approach offers a potentially powerful means

to identify both new targets of Beps and their FIC

domain-mediated post-translational modifications.

Current Opinion in Microbiology 2015, 23:80–85

Concluding remarksRecent advances have enhanced our understanding of

how the repertoire of Beps finely manipulates host cell

processing steps enabling Bartonella to interact, colonize,

proliferate and persist in host cells. Here we have

reviewed the evolutionary diversification of these effector

proteins, their function and their phenotypic properties.

Although they share a modular architecture, they

acquired different functions within host cells. Some

effectors, like BepC, BepF and BepG, appear to act in

concert whereas others, like BepC and BepE, appear to

be antagonistic. However, despite the progress made to

elucidate the functions of several Beps from lineage 4, the

roles of the ones from the lineage 3 remain to be elucidated.

One of the big challenges in the study of bacterial effectors

is that they exert their function(s) in concert with other

effectors and therefore are regulated both temporally and

spatially. We hope that new approaches developed will

enhance our understanding of the contribution of these

proteins to the Bartonella–host interaction, making the

next few years very exciting in this regard.

Acknowledgements

We would like to thank Alexander Harms, Dr. Rusudan Okujava andDr. Maxime Quebatte for critical reading of the manuscript. This work wassupported by Grant 310030B_149886 from the Swiss National ScienceFoundation (SNSF, www.snf.ch), advanced Grant 340330 (FicModFun)from the European Research Council (ERC), and Grant 51RTP0_151029for the research and Technology Development (RTD) projectTargetInfectX in the frame of systemsX.ch (www.systemX.ch), the SwissInitiative for System Biology.

References and recommended readingPapers of particular interest, published within the period of review,have been highlighted as:

� of special interest�� of outstanding interest

1. Chomel BB, Boulouis HJ, Breitschwerdt EB, Kasten RW, Vayssier-Taussat M, Birtles RJ, Koehler JE, Dehio C: Ecological fitnessand strategies of adaptation of Bartonella species to theirhosts and vectors. Vet Res 2009, 40:29.

www.sciencedirect.com

Function of Bartonella effector proteins Siamer and Dehio 85

2.��

Okujava R, Guye P, Lu YY, Mistl C, Polus F, Vayssier-Taussat M,Halin C, Rolink AG, Dehio C: A translocated effector required forBartonella dissemination from derma to blood safeguardsmigratory host cells from damage by co-translocatedeffectors. PLoS Pathog 2014, 10:e1004187.

Identified BepE as an essential protein for Bartonella dissemination fromthe dermal site of inoculation to the blood stream.

3. Pulliainen AT, Dehio C: Persistence of Bartonella spp. stealthpathogens: from subclinical infections to vasoproliferativetumor formation. FEMS Microbiol Rev 2012, 36:563-599.

4.�

Angelakis E, Raoult D: Pathogenicity and treatment ofBartonella infections. Int J Antimicrob Agents 2014, 44:16-25.

A comprehensive review covering the clinical aspect of Bartonella infec-tions.

5.�

Harms A, Dehio C: Intruders below the radar: molecularpathogenesis of Bartonella spp.. Clin Microbiol Rev 2012,25:42-78.

A comprehensive review covering the molecular and cellular basis ofpathogenesis by Bartonella spp.

6. Engel P, Salzburger W, Liesch M, Chang CC, Maruyama S, Lanz C,Calteau A, Lajus A, Medigue C, Schuster SC, Dehio C: Parallelevolution of a type IV secretion system in radiating lineages ofthe host-restricted bacterial pathogen Bartonella. PLoS Genet2011, 7:e1001296.

7.�

Guy L, Nystedt B, Toft C, Zaremba-Niedzwiedzka K, Berglund EC,Granberg F, Naslund K, Eriksson AS, Andersson SG: A genetransfer agent and a dynamic repertoire of secretion systemshold the keys to the explosive radiation of the emergingpathogen Bartonella. PLoS Genet 2013, 9:e1003393.

A study which proposed an updated Bartonella phylogeny based on thesequencing of six new genomes and description of the highly conservedGTA in Bartonella genomes.

8.��

Zhu Q, Kosoy M, Olival KJ, Dittmar K: Horizontal transfers andgene losses in the phospholipid pathway of Bartonella revealclues about early ecological niches. Genome Biol Evol 2014,6:2156-2169.

A new phylogenetic study highlighting the impact of horizontal genetransfer on Bartonellae lineage evolution.

9. Saenz HL, Engel P, Stoeckli MC, Lanz C, Raddatz G, Vayssier-Taussat M, Birtles R, Schuster SC, Dehio C: Genomic analysis ofBartonella identifies type IV secretion systems as hostadaptability factors. Nat Genet 2007, 39:1469-1476.

10. Vayssier-Taussat M, Le Rhun D, Deng HK, Biville F, Cescau S,Danchin A, Marignac G, Lenaour E, Boulouis HJ, Mavris M et al.:The Trw type IV secretion system of Bartonella mediateshost-specific adhesion to erythrocytes. PLoS Pathog 2010,6:e1000946.

11. Schulein R, Guye P, Rhomberg TA, Schmid MC, Schroder G,Vergunst AC, Carena I, Dehio C: A bipartite signal mediates thetransfer of type IV secretion substrates of Bartonella henselaeinto human cells. Proc Natl Acad Sci USA 2005, 102:856-861.

12. Cruz JW, Rothenbacher FP, Maehigashi T, Lane WS, Dunham CM,Woychik NA: Doc toxin is a kinase that inactivates elongationfactor tu. J Biol Chem 2014, 289:7788-7798.

This study demonstrates the phosphorylation activity of the Doc toxinwhich represents a new catalytic activity for the FIC protein family.

13. Engel P, Goepfert A, Stanger FV, Harms A, Schmidt A, Schirmer T,Dehio C: Adenylylation control by intra- or intermolecularactive-site obstruction in FIC proteins. Nature 2012,482:107-110.

A structure–function study reporting that the Bep ancestral protein VbhTof B. schoenbuchensis forms a toxin–antitoxin module together with thesmall VbhA protein.

14. Mukherjee S, Liu X, Arasaki K, McDonough J, Galan JE, Roy CR:Modulation of Rab GTPase function by a proteinphosphocholine transferase. Nature 2011, 477:103-106.

www.sciencedirect.com

15. Worby CA, Mattoo S, Kruger RP, Corbeil LB, Koller A, Mendez JC,Zekarias B, Lazar C, Dixon JE: The FIC domain: regulation of cellsignaling by adenylylation. Mol Cell 2009, 34:93-103.

16. Berglund EC, Frank AC, Calteau A, Vinnere Pettersson O,Granberg F, Eriksson AS, Naslund K, Holmberg M, Lindroos H,Andersson SG: Run-off replication of host-adaptability genesis associated with gene transfer agents in the genome ofmouse-infecting Bartonella grahamii. PLoS Genet 2009,5:e1000546.

17. Stavrinides J, Ma W, Guttman DS: Terminal reassortment drivesthe quantum evolution of type III effectors in bacterialpathogens. PLoS Pathog 2006, 2:e104.

18.�

Lang AS, Zhaxybayeva O, Beatty JT: Gene transfer agents:phage-like elements of genetic exchange. Nat Rev Microbiol2012, 10:472-482.

A comprehensive review covering the characteristics of GTAs.

19. Dehio C: Infection-associated type IV secretion systems ofBartonella and their diverse roles in host cell interaction.Cell Microbiol 2008, 10:1591-1598.

20. Eicher SC, Dehio C: Bartonella entry mechanisms intomammalian host cells. Cell Microbiol 2012, 14:1166-1173.

21. Rhomberg TA, Truttmann MC, Guye P, Ellner Y, Dehio C: Atranslocated protein of Bartonella henselae interferes withendocytic uptake of individual bacteria and triggers uptake oflarge bacterial aggregates via the invasome. Cell Microbiol2009, 11:927-945.

22. Truttmann MC, Rhomberg TA, Dehio C: Combined actionof the type IV secretion effector proteins BepC andBepF promotes invasome formation of Bartonella henselaeon endothelial and epithelial cells. Cell Microbiol 2011,13:284-299.

23. Truttmann MC, Guye P, Dehio C: BID-F1 and BID-F2 domains ofBartonella henselae effector protein BepF trigger togetherwith BepC the formation of invasome structures. PLoS One2011, 6:e25106.

24.��

Pulliainen AT, Pieles K, Brand CS, Hauert B, Bohm A, Quebatte M,Wepf A, Gstaiger M, Aebersold R, Dessauer CW, Dehio C:Bacterial effector binds host cell adenylyl cyclase topotentiate GaS-dependent cAMP production. Proc Natl AcadSci USA 2012, 109:9581-9586.

This functional study reveals host adenylate cyclase as a target for theanti-apoptotic activity of BepA.

25. Schmid MC, Scheidegger F, Dehio M, Balmelle-Devaux N,Schulein R, Guye P, Chennakesava CS, Biedermann B, Dehio C: Atranslocated bacterial protein protects vascular endothelialcells from apoptosis. PLoS Pathog 2006, 2:e115.

26. Palanivelu DV, Goepfert A, Meury M, Guye P, Dehio C, Schirmer T:FIC domain-catalyzed adenylylation: insight provided by thestructural analysis of the type IV secretion system effectorBepA. Protein Sci 2011, 20:492-499.

27. Worthylake RA, Lemoine S, Watson JM, Burridge K: RhoA isrequired for monocyte tail retraction during transendothelialmigration. J Cell Biol 2001, 154:147-160.

28. Selbach M, Paul FE, Brandt S, Guye P, Daumke O, Backert S,Dehio C, Mann M: Host cell interactome of tyrosine-phosphorylated bacterial proteins. Cell Host Microbe 2009,5:397-403.

29.��

Pieles K, Glatter T, Harms A, Schmidt A, Dehio C: An experimentalstrategy for the identification of AMPylation targetsfrom complex protein samples. Proteomics 2014,14:1048-1052.

New proteomic approach to study post-translational modifications ofhost proteins by bacterial effectors.

Current Opinion in Microbiology 2015, 23:80–85