Cellular Microbiology (2001) 3(10), 693±702

Yersinia enterocolitica invasin triggers phagocytosis viab1 integrins, CDC42Hs and WASp in macrophages

AgneÁs Wiedemann,1 Stefan Linder,1,2 Guntram

Grassl,4 Michael Albert,3 Ingo Autenrieth4 and Martin

Aepfelbacher1*1Max von Pettenkofer-Institut fuÈr Medizinische

Mikrobiologie, Pettenkoferstrasse 9a and 2Institut fuÈr

Prophylaxe und Epidemiologie der Kreislaufkrankheiten,

Pettenkoferstrasse 9, Ludwig-Maximilians-UniversitaÈ t,

80336 MuÈnchen, Germany. 3Dr von Haunersches

Kinderspital, Klinikum Innenstadt, Lindwurmstr. 4, 80337

MuÈnchen, Germany. 4Institut fuÈr Medizinische

Mikrobiologie, UniversitaÈ t TuÈbingen, Germany.

Summary

The Yersinia outer surface protein invasin binds to b1

integrins on target cells and has been shown to

trigger phagocytic uptake by macrophages. Here, we

investigated the role of the actin regulator Wiskott±

Aldrich syndrome protein (WASp), its effector the

Arp2/3 complex and the Rho-GTPases CDC42Hs, Rac

and Rho in invasin/b1 integrin-triggered phagocyto-

sis. During uptake of invasin-coated latex beads, the

a5b1 integrin, WASp and the Arp2/3 complex were

recruited to the developing actin-rich phagocytic

cups in primary human macrophages. Blockage of b1

integrins by specific antibodies, inhibition of Arp2/3

function by microinjection of inhibitors or the use of

WASp knockout macrophages inhibited phagocytic

cup formation and uptake. Furthermore, microinjec-

tion of the dominant negative GTPase mutants

N17CDC42Hs, N17Rac or the Rho-specific inhibitor

C3-transferase into macrophages greatly attenuated

invasin-induced formation of cups. These data sug-

gest that during invasin-triggered phagocytosis b1

integrins activate actin polymerization via CDC42Hs,

its effector WASp and the Arp2/3 complex. The

contribution of Rac and Rho to phagocytic cup

formation also suggests a complex interplay

between different Rho GTPases during phagocytosis

of pathogens.

Introduction

Yersinia enterocolitica and Yersinia pseudotuberculosis,

pathogenic bacteria that cause food borne acute or

chronic gastroenteric diseases, express on their surface

the b1 integrin-binding protein invasin (Isberg and Leong,

1990; Young et al., 1990). The invasins of the two

enteropathogenic Yersinia species are highly homologous

except for a self-association region present only in Y.

pseudotuberculosis invasin (Dersch and Isberg, 1999).

The region of invasin necessary and sufficient for integrin

binding was mapped to the C-terminal 192 amino acids

and was shown to contain a motif resembling the

consensus RGD motif of integrin ligands (Leong et al.,

1995). Invasin is thought to mediate Yersinia invasion into

and translocation through the M-cells of the intestinal

mucosa, which overlay specialized lymph follicles called

Peyer's patches (Autenrieth and Firsching, 1996; Marra

and Isberg, 1997). Interestingly, invasin may also trigger

uptake of yersiniae by phagocytic immune cells located

within the Peyer's patches (Pepe et al., 1995; Cornelis

and Wolf-Watz, 1997). Consistent with this idea, it was

demonstrated that invasin mediates phagocytosis of Y.

pseudotuberculosis by cultured J774 macrophages in the

absence of opsonizing antibodies (FaÈ llman et al., 1995).

Yersinia spp. are able to block their own phagocytic

uptake and, like many other pathogens, can modulate

macrophage responses to subvert the host immune

system (Cornelis and Wolf-Watz, 1997; Finlay and

Cossart, 1997). Which signal pathways Yersinia must

act on to block phagocytosis or to modulate macrophage

responses probably depends on the type of surface

receptor mediating the uptake. Among the best investi-

gated phagocytic receptors are complement receptors

and Fc-receptors that bind to the respective opsonized

particles as well as mannose-binding receptors and

lipopolysaccharide-binding scavenger receptors that

directly associate with conserved motifs on the pathogens

(Ofek et al., 1995; Silverstein, 1995; Vidarsson and van

de Winkel, 1998). As exemplified by Yersinia, phagocytic

uptake can also be mediated by integrins normally

involved in extracellular matrix adhesion, such as the

a5b1 fibronectin receptor (Pommier et al., 1983; Wright

et al., 1983; Blystone et al., 1994). Considering that,

besides Yersinia, a variety of other pathogens express

integrin-binding adhesins (Virji, 1996), elucidation of the

signal pathways involved in integrin dependent phagocy-

tosis is crucial to understand how pathogenic bacteria

modulate immune responses.

Phagocytosis is controlled by complex rearrangements

Q 2001 Blackwell Science Ltd

Received 22 January, 2001; revised 17 July, 2001; accepted 19 July,2001. *For correspondence. E-mail aepfelbacher@m3401.mpk.med.uni-muenchen.de; Tel. (149) 89 5160 5264; Fax (149) 89 51605223.

of the actin cytoskeleton, which is also reflected by the

great number of actin-regulatory proteins that localize to

phagocytic uptake structures (Massol et al., 1998; Aderem

and Underhill, 1999; Allen and Aderem, 1995). Recently,

the GTP-binding protein Rho was found to be involved in

complement receptor (CR3) dependent phagocytosis,

whereas Rac and CDC42Hs have been shown to

specifically regulate the assembly of actin-rich phagocytic

cups during FcgR-mediated phagocytosis (Cox et al.,

1997; Caron and Hall, 1998; Massol et al., 1998).

According to the current model, Rho reorganizes pre-

formed actin filaments and actomyosin contraction

(Machesky and Hall, 1997), whereas CDC42Hs and Rac

induce de novo actin polymerization via effector proteins of

the Wiskott±Aldrich syndrome protein (WASp) family,

which includes WASp, N-WASp as well as human Scar/

WAVE (Higgs and Pollard, 1999). Upon binding to active,

GTP-bound CDC42Hs, WASp is thought to expose its C-

terminal domain, which then greatly enhance the actin

polymerizing activity of the Arp2/3 complex (Higgs and

Pollard, 1999; Kim et al., 2000). The Arp2/3 complex

consists of seven polypeptides and is able to do both, de

novo nucleate actin filaments and elongate actin filaments

at their barbed ends (Higgs and Pollard, 1999). Given the

central role of Arp2/3 in actin polymerization, it is not

surprising that it is also required for FcgR-mediated actin

polymerization and phagocytosis (May et al., 2000). The

upstream regulators that control Arp2/3 in specific phago-

cytes may be diverse but, in blood monocytes of WAS

patients, formation of FcgR-induced phagocytic cups was

attenuated (Lorenzi et al., 2000).

We report here that Yersinia invasin binding to b1

integrins triggers formation of phagocytic cups that are

dependent on WASp, the Arp2/3 complex and the

GTPases CDC42Hs, Rac and Rho. These results identify

Rho GTPases, WASp and the Arp2±3 complex as major

regulators of b1 integrin-mediated actin reorganisation

leading to phagocytic uptake of pathogens.

Results

Yersinia invasin triggered phagocytosis is preceded by

formation of actin-rich phagocytic cups

Because macrophages can phagocytose and polymerize

actin into phagocytic cups even in response to inert latex

beads, we first sought to clarify the role of Yersinia invasin as

a trigger for phagocytic uptake. For this purpose, the

following sets of bacteria/beads either exposing invasin or

not were tested: (i) plasmid-free Yersinia enterocolitica (WA-

C), pre-cultured at 378C or 278C to downregulate or allow

expression of invasin respectively; (II) Escherichia coli

HB101 or E. coli HB101 inv1, the latter expressing Y.

enterocolitica invasin (Schulte et al., 2000); and (iii) latex

beads (2- or 4-mm-diameter) coated with glutathione-S-

transferase (GST) alone or with GST-fusion proteins

containing the C-terminal 397 (GST±Inv397) or 195

(GST±Inv195) amino acids of Y. enterocolitica invasin.

Figure 1 demonstrates the absence or presence, respec-

tively, of invasin on these particles. Beads/bacteria were

centrifuged onto human macrophages at 48C followed by a

15 min feeding period at 378C and actin staining with

rhodamine phalloidin. Only beads that were fully surrounded

by an actin ring were scored as phagocytic cup-positive. As

shown in Fig. 1, bacteria or beads expressing invasin

induced up to 10-fold more phagocytic cups per cell than

their invasin-free counterparts. Only a few per cent of the

total macrophage-attached bacteria showed phagocytic

cups, whereas about 33% of the adherent GST±Inv397-

coatedbeads producedcups (Fig. 1). GST±Inv195 was less

Fig. 1. Yersinia invasin triggers formation of phagocytic cups in human macrophages. GST- or GST±Inv397-coated latex beads (2 mm-diameter), Yersinia enterocolitica (WA-C) pre-cultured at 378C or 278C, E. coli HB101 or E. coli HB101 inv 1 were tested for expression ofinvasin using Western blot analysis (IB). Full length invasin and the GST-fused truncated form (GST±Inv397) run at 107 kDa and 70 kDarespectively. For visualization of phagocytic cups, primary human macrophages were infected for 15 min at 378C with bacteria/beads at acalculated particle:cell ratio of 50±100:1 and stained for actin using rhodamine phalloidin. Only actin accumulations that formed a continuousring around the beads/bacteria were scored as phagocytic cup-positive (for example see Fig. 3A and B). The total number of adherentparticles was determined using phase contrast microscopy. Values are mean of three experiments in which at least 33 cells were evaluatedper data point. Standard deviations were usually within 20±30% of values.

694 A. Wiedemann et al.

Q 2001 Blackwell Science Ltd, Cellular Microbiology, 3, 693±702

efficient in phagocytic cup formation than GST±Inv397 (see

Fig. 4). Also, because actin accumulations could best be

visualized with GST±Inv397-coated beads (Fig. 3A and B),

these particles were used for the remaining experiments.

Next, we tested how actin cup formation is related to

phagocytic uptake in a time course experiment. An inside-

outside fluorescence technique revealed that GST±

Inv397-coated beads were rapidly taken up by the

macrophages and, after 60±90 min, all macrophage-

associated beads (<25 per cell) were intracellular. Most

phagocytic cups were seen at the maximal uptake rate

(5±15 min) and, thereafter, their number decreased

rapidly (Fig. 2). These experiments indicate that actin

cups are formed during a certain window of invasin-

triggered phagocytic uptake. We propose that this window

is longer for the relatively big invasin-coated beads than

for invasin expressing bacteria, which could explain the

much higher number of phagocytic cups seen with the

beads versus the bacteria.

Yersinia invasin-triggered phagocytosis requires b1

integrins

To determine which integrin receptors normally expressed

by macrophages may be involved in invasin-induced

phagocytosis, we first performed double fluorescence

stainings using anti-invasin antibody and antibodies either

against the b1-, b2- or av-subunits of integrins or the

a5b1-, avb5 or avb3 integrin receptors. We found that

within phagocytic cups, invasin co-localized with b1 and

b2 integrins as well as with the a5b1 integrin receptor

but not with the avb5 or avb3-integrin receptors or

other av-containing integrins (Fig. 3C±E and data not

shown). Figure 3(C±E) demonstrates that invasin bound

to the surface of the beads and the a5b1 integrin

receptor, expressed on the macrophage plasma mem-

brane, co-localize within phagocytic cups. Furthermore,

pre-treatment of macrophages with anti-b1 integrin anti-

body or RGD (Arg-Gly-Asp) or RGDS (RGD-Ser) peptides

Fig. 2. Time relation between phagocytic cup formation andinternalization of Inv-beads. GST±Inv397-coated beads (4-mm-diameter) were centrifuged onto macrophages at 48C at aparticle:cell ratio of 25:1. At 0 min, the temperature was raised to378C and at indicated time points thereafter the number ofphagocytic cups and intracellular beads was determined as in theExperimental procedures. Values are mean ^SD of threeexperiments. At least three times 30 cells were evaluated perexperiment.

Fig. 3. Actin and the a5b1 integrin receptoraccumulate in invasin-triggered phagocyticcups. Macrophages were infected for 15 minat 378C with GST±Inv397-coated beads(4 mm-diameter) at a calculated bead:cell ratioof 25:1 and then stained for actin, invasin ora5b1 integrin as in the Experimentalprocedures. Horizontal sections of apical partsof cells were recorded using confocal laserscanning microscopy.A. Actin staining in red.B. Phase contrast picture of the cell in A tovisualize beads.C. Invasin staining in green.D. a5b1 Integrin staining in red.E. Overlay of (C) and (D), yellow colourindicates co-localization of green and red;(Bar � 10 mm).

Yersina invasion-induced phagocytosis is regulated by WASp 695

Q 2001 Blackwell Science Ltd, Cellular Microbiology, 3, 693±702

reduced the number of phagocytic cups triggered by the

GST-Inv beads to numbers similar to those obtained with

GST-coated beads (Fig. 4A). Similarly, internalization of

the Inv-beads was reduced drastically by the anti-b1

antibodies and RGD peptide (Fig. 4B). Together these

data suggest that invasin-triggered actin polymerization

and phagocytosis depend on b1 integrins.

WASp and the Arp2/3 complex are involved in Yersinia

invasin-triggered phagocytosis.

To test whether WASp and the Arp2/3 complex are

involved in invasin-triggered phagocytosis, we first per-

formed double fluorescence stainings using an antibody

against p41-Arc, a subunit of the Arp2/3 complex (Higgs

and Pollard, 1999) and either rhodamine phalloidin or anti-

WASp antibody. These experiments indicated co-locali-

zation of actin, Arp2/3 and WASp within invasin-induced

phagocytic cups (Fig. 5A±F).

To obtain unambiguous evidence that WASp regulates

formation of phagocytic cups triggered by invasin, we

investigated macrophages from a WAS patient with a

C995T mutation in the WASp gene leading to a premature

stop codon. In cells of this patient, no WASp could be

detected at the protein or mRNA level (Linder et al.,

1999). As tested by actin- and Arp2/3 staining, formation

of phagocytic cups was reduced drastically in the WAS

macrophages (Fig. 6A and B and Fig. 7A for quantita-

tion). Surprisingly, we found that internalization of beads

was inhibited by only 30±40% in the WAS cells (Fig. 7B).

We also microinjected into normal macrophages two

GST-fused C-terminal fragment of WASp, named GST-A

and GST-VC, which are thought to block interaction of

WASp with the Arp2/3 complex (Linder et al., 2000). Both

of these constructs greatly inhibited invasin-triggered cup

formation and phagocytic uptake of beads (Fig. 7A and B

and not shown) and of yersiniae expressing invasin (data

not shown).

CDC42Hs, Rac and Rho are required for Yersinia invasin-

induced cup formation.

The activity of the RhoGTPase CDC42Hs can be

modulated by integrins (Price et al., 1998; Machesky

and Hall, 1997) and, in turn, CDC42Hs can control the

activity of WASp (Higgs and Pollard, 1999). To test

whether CDC42Hs may also be involved in invasin-

triggered phagocytic cup formation, we microinjected

dominant negative N17CDC42Hs into macrophages. We

found that in the N17CDC42 microinjected cells, formation

of phagocytic cups around invasin-coated beads was

reduced drastically (Fig. 6C and D and Fig. 8). Further

experiments revealed that microinjection of dominant

negative N17Rac or the isolated GTPase binding domain

of CDC42Hs/Rac target PAK (Daniels and Bokoch, 1999)

as well as the Rho-specific inhibitor C3-transferase

(Aepfelbacher et al., 1996) or the isolated Rho-binding

domain of the Rho target Rho-kinase (RBD) (Essler et al.,

1998) blocked cup formation by beads (Fig. 8). Similar

results were obtained with yersiniae expressing invasin

(data not shown).

Altogether, these results suggest that Rho GTPases,

WASp and the actin regulatory Arp2/3 complex are crucial

for invasin/b1 integrin-triggered actin polymerization and

phagocytosis in macrophages.

Discussion

Although a variety of bacterial pathogens can associate

with leukocytic b1 integrins via their surface adhesins

(Virji, 1996), very limited data have been available on the

role of this interaction in phagocytosis. Earlier studies

showed that attachment of macrophages to fibronectin

(via the a5b1 integrin receptor) stimulated C3bi-mediated

phagocytic uptake (through the complement receptor

CR3; aMb2; CD11b/CD18a) (Pommier et al., 1983;

Wright et al., 1983). Thus, the b2 integrins detected in

the invasin-induced phagocytic cups may represent

complement receptors and may thus indicate a

Fig. 4. Invasin-triggered phagocytic cup formation andphagocytosis is dependent on b1 integrins. Macrophages were nottreated (Ctrl) or pre-incubated for 15 min at 378C with anti-b1antibodies (1:10), RGD or RGDS peptides (1 mM each) asindicated. Thereafter, macrophages were infected for 15 min at378C with GST-, GST±Inv195- or GST±Inv397-coated latex beads(2 mm-diameter) at a calculated bead:cell ratio of 50:1. The numberof (A) phagocytic cups and (B) intracellular beads per cell wasdetermined as in the Experimental procedures. An unspecificmouse IgG antibody had no effect. Each value represents mean^ SD of three different experiments with at least 33 cells evaluatedper experiment.

696 A. Wiedemann et al.

Q 2001 Blackwell Science Ltd, Cellular Microbiology, 3, 693±702

co-operation of integrin receptors and complement

receptors in invasin/b1 integrin-triggered phagocytosis.

An important point concerns the signalling mechanisms

by which b1 integrins activate actin reorganisation during

phagocytosis. The Rho GTPases CDC42Hs, Rac and

Rho have all been shown to be activated by ligation of b1

integrin receptors in different experimental settings,

including cell spreading on extracellular matrices

(Machesky and Hall, 1997; Bourdoulous et al., 1998;

Price et al., 1998; Ren et al., 1999; Mainiero et al., 2000).

The finding that CDC42Hs, Rac and Rho are also required

for invasin-triggered formation of phagocytic cups indi-

cates once more the similarities in the molecular actin

organization of cell spreading/migration and phagocyto-

sis. The requirement for the CDC42Hs-specific effector

WASp in invasin-stimulated cup formation suggests that

binding of invasin to macrophage b1 integrins activates

CDC42Hs, which then stimulates WASp and the Arp2/3

complex. However, it can not be excluded that WASp is

recruited directly to integrins via binding to adapter

proteins such as Grb2 or NCK or via non-receptor

tyrosine kinases such as Fyn (Hy et al., 1997; Nonoyama

and Ochs, 1998; Rivero-Lezcano et al., 1998 Giancotti,

2000). How Rac and Rho contribute to invasin-induced

phagocytosis is also unclear at the moment, but Rac

seems to regulate actin through the WASp family proteins

WAVE/Scar (1±3) and Rho controls contractility of

actomyosin via its target Rho-kinase in many cell types

(Higgs and Pollard, 1999; Bishop and Hall, 2000). Rho

and Rac can also inhibit actin depolymerization via LIM-

kinase-mediated phosphorylation and inactivation of

cofilin (Yang et al., 1998; Sumi et al., 1999).

The finding that internalization of invasin-coated beads

was reduced by only 30±40% in WAS macrophages

indicates WASp-independent mechanisms of phagocyto-

sis. Others have found that phagocytosis of apoptotic

cells is reduced by 30±40%, and formation of actin-rich

phagocytic cups is diminished but not abolished in WAS

monocytes (Leverrier et al., 2001). Therefore, it is

likely that WASp family proteins such as WAVE/Scar or

N-WASp can partially take over WASp function in the

Wiskott±Aldrich syndrome. The highly efficient inhibition

of cup formation and uptake by the VC- and A-domain of

WASp, which are thought to act as general inhibitors of

WASp family proteins and the Arp2/3 complex, further

support this notion.

A well known feature of Yersinia spp. is to inject effector

proteins, so called Yops (Yersinia outer proteins), into the

Fig. 5. Actin, Arp2±3 complex and WASp co-localize in phagocytic cups of human macrophages. Macrophages were infected for 15 min at378C with GST±Inv397-coated beads (4-mm-diameter) at a calculated bead:cell ratio of 25:1 and then stained for actin, Arp2±3 complex andWASp as in the Experimental procedures. Horizontal sections of apical parts of cells were recorded using confocal laser scanning microscopy.A. Actin staining in red.B. Arp2/3 complex staining in blue.C Overlay of (A) and (B), pink colour indicates co-localization of red and green.D. WASp staining in red.E. Arp2/3 complex staining in blue.F. Overlay of (D) and (E), pink colour indicates co-localization of red and blue; (Bar � 10 mm).

Yersina invasion-induced phagocytosis is regulated by WASp 697

Q 2001 Blackwell Science Ltd, Cellular Microbiology, 3, 693±702

cytosol of macrophages by making use of a type III

secretion system (Cornelis and Wolf-Watz, 1997). The

Yops are highly adapted proteins that specifically block

macrophage signal transduction pathways involved in

antiapoptosis, cytokine release or phagocytosis (Aepfel-

bacher et al., 1999). Remarkably, at least three Yops ±

YopE, YopT and YopO (Yersinia enterocolitica nomen-

clature) ± impinge on Rho GTPase-regulated signalling

events. YopE acts as a GTPase-activating protein (GAP)

for different Rho GTPases in vitro and specifically

inactivates Rac in endothelial cells (von Pawel-Rammin-

gen et al., 2000; Andor et al., 2001). Recently, YopT was

found to modify and inactivate the small GTP-binding

protein RhoA (Zumbihl et al., 1999), whereas YopO binds

directly to Rac-1 and RhoA and may thereby influence

their function (Barz et al., 2000; Dukuzumuremyi et al.,

2000). Furthermore, another Yop, YopH, is a tyrosine

phosphatase that dephosphorylates focal adhesion

kinase and p130Cas, two signalling scaffolds regulating

the association of integrins with the actin cytoskeleton

(Black and Bliska, 1997; Persson et al., 1997). Hence,

downregulation of Rho GTPase-mediated signalling

events, such as those initiated during invasin/b1 integ-

rin-induced phagocytosis, seems to be crucially important

for Yersinia to successfully establish infection.

In summary, we show here that invasin of Yersinia

enterocolitica induces the formation of phagocytic cups in

human macrophages by a b1 integrin-mediated pathway, in

which Rho GTPases, WASp and the Arp2±3 complex are

crucially involved. Thus, integrin signalling during phagocy-

tosis leads to recruitment of a complex actin polymerizing

machinery, which may be a common endpoint of other

Fig. 6. Reduced phagocytic cup formation in WAS macrophagesand macrophages microinjected with N17CDC42Hs. WASmacrophages, normal macrophages or N17CDC42Hs-microinjectedmacrophages were infected for 15 min at 378C with GST±Inv397-coated beads (4 mm-diameter) at a calculated bead:cell ratio of25±50:1. Horizontal sections of apical parts of cells were recordedby confocal laser scanning microscopy.A. Actin staining of a WAS macrophage.B. Phase contrast of the cell in A to visualize beads.C. Actin staining of a N17CDC42Hs microinjected macrophage (leftlower corner) surrounded by normal macrophages.D. FITC-dextran fluorescence identifies the microinjected cell of (C);(Bar � 10 mm).

Fig. 7. Phagocytic cup formation is dependent on WASp andArp2/3 complex. Macrophages were not injected, microinjected withFITC-Dextran, GST-A or GST-VC or from a WAS patient (WAScells). Cells were infected for 15 min at 378C with GST±Inv397-coated latex beads (2 or 4 mm-diameter) at a calculated beads:cellratio of 50±100:1 and (A) the number of phagocytic cups and totalcell associated beads (adherent beads) or (B) intracellular beadswas determined as in the Experimental procedures. Each valuerepresents mean ^ SD of three different experiments with at leastthree times 20 cells evaluated per experiment.

Fig. 8. Phagocytic cup formation depends on CDC42Hs, Rac andRho. Macrophages were not injected or microinjected with FITC-Dextran, C3-transferase (C3), Rho-binding domain of Rho-Kinase(RBD), N17Rac, CRIB domain of CDC42Hs/Rac target PAK (PAK)or N17CDC42Hs, incubated for 30 min and then infected for 15 minat 378C with GST±Inv397-coated latex beads (4 mm-diameter) at acalculated beads:cell ratio of 50:1. The number of phagocytic cupsper cell was determined as in the Experimental procedures. Eachvalue represents mean ^SD of three different experiments with atleast three times 20 cells evaluated per experiment.

698 A. Wiedemann et al.

Q 2001 Blackwell Science Ltd, Cellular Microbiology, 3, 693±702

integrin-stimulated events such as cell migration, cell shape

control or cell proliferation.

Experimental procedures

Bacterial strains and growth conditions

Plasmid-free Yersinia enterocolitica (WA-C), non-invasive E.coli HB101 (inv-), and the E. coli HB101 inv1 strainexpressing the Y. enterocolitica Inv (Schulte et al., 1998)were grown in Luria±Bertani broth (LB) with antibioticselection. For infection experiments, overnight cultureswere grown at 278C and then diluted (1:10) in LB and furtherincubated for 2 h at 278C or 378C respectively.

Cell isolation and cell culture

Human peripheral blood monocytes were isolated bycentrifugation of heparinized blood in Ficoll (Seromed).Monocytes were isolated with magnetic anti-CD14 antibodybeads and an MS1 Separation Column (Miltenyi Biotec)according to the manufacturer's instructions and seeded ontoCellocate coverslips (Eppendorf) at a density of 5 � 104

cells. Cells were cultured for 5±7 days in RPMI containing20% autologous serum at 378C, 5% CO2 and 90% humidity.The medium was changed every 3±4 days. Within 5±7 days,monocytes differentiate into adherent, spread out macro-phages (Linder et al., 1999). WAS macrophages wereisolated and cultured identically. Cells from a WAS patienthave been used for this study with the explicit permission ofthe patient.

DNA constructs

Part of the inv gene encoding the 397 and 195 C-terminalamino acids (termed Inv397 and Inv195 respectively) wasamplified from Y. enterocolitica serotype O:8 by polymerasechain reaction (PCR) using the forward primer 5 0-ACGTGAATTCCACGTTGACCGTTATTGTGC-3 0 (EcoRI restric-tion site underlined) for Inv397, 5 0-ACGTGAATTCCTACCCAGTACCGAAGATAA-3 0 for Inv195, and the reverse primer 5 0-GCCGCTCGAGCTATTGCGGCTCCGCAC-3 0 (XhoI restric-tion site underlined) according to the published sequence(Young et al., 1990). The amplified DNA fragments (digestedwith EcoRI and XhoI) were cloned into pGEX-4T-3 expres-sion vector (Amersham-Pharmacia) resulting in pINV397 orpINV195.

Expression and purification of GST±Inv397 and

GST±Inv195 fusion protein

Escherichia coli BL21, harbouring pGEX-4T-3-pINV397, wasgrown at 248C in LB to an optical density of 0.7. Expression ofthe GST- and GST-Inv fusion proteins was induced withIPTG (isopropyl-b-D-thiogalactopyranoside) at a final con-centration of 0.1 mM. Cells were grown for an additional 2 hbefore being harvested by centrifugation and frozen at2208C. Frozen cells were resuspended in phosphate-buffered saline (PBS) containing 1 mM phenylmethylsulfonyl-fluoride (PMSF) and complete protease inhibitor cocktail(Boehringer). After disrupting bacterial cells using a Frenchpress, lysates were cleared for cellular debris by centrifugation.

GST±Inv397, GST±Inv195 and GST were purified fromsupernatants using Glutathion Sepharose 4B (PharmaciaBiotech) and Superdex 200 gel filtration. Purity and identity ofGST±Inv397 and GST±Inv195 fusion proteins were ana-lysed by SDS±PAGE and Western blot using goat anti-GST(Pharmacia; diluted 1:1000) and anti-Inv 3.1 antisera (diluted1:5000) (Schulte et al., 1998). Protein concentration wasdetermined using the BCA protein assay (Pierce).

Coating of proteins to latex beads

For non-covalent coating of beads, purified protein wasdialyzed against PBS pH 7.0. About 109 latex beads (2-mm-diameter, polystyrene sulphate-modified, Sigma; for someexperiments 4-mm-diameter beads, sulphate microspheres,Molecular Probes, were used) were washed with 1 ml of PBSand resuspended in 500 ml of PBS. Purified GST, GST±Inv397 or GST±Inv195 fusion proteins (1 mg) were addedand allowed to adsorb to the beads for 3 h at roomtemperature (RT). After adding 500 ml of 20 mg ml21 BSA,the solution was incubated at RT for another 1 h. Next, beadswere washed in PBS containing 1 mg ml21 BSA and storedat 48C in 500 ml of PBS containing 0.2 mg ml21 BSA. Todetermine the coupling efficiency, the protein concentrationof the starting solution and of the supernatant before addingBSA was determined. Integrity of coated GST±Inv397, GST±Inv195 and GST fusion protein was checked using Westernblot analysis.

SDS±PAGE and Western blot analysis

Bacterial strains, GST- or GST-Inv397 beads were lysed inSDS sample buffer (about 1.5 � 107 E. coli, 9 � 107 Yersiniabacteria or 2.5 � 106 beads were loaded per lane). Separa-tion of the proteins was performed using SDS±PAGE. Forimmunoblotting, the proteins were transferred electrophor-etically onto polyvinylene difluoride (PVDF) membranes (1,2mA/cm2 for 60min). The sheets were blocked for 1 h withPBS/1% BSA/5% milk at room temperature. For detection ofinvasin, a polyclonal rabbit anti-Inv 3.1 antiserum (diluted1:5000) and a peroxidase-conjugated secondary anti-rabbitantibody (Amersham; diluted 1:10000) were added for 1 heach. The detection was carried out using the chemilumines-cence detection kit (ECL) from Amersham.

Immunofluorescence microscopy of phagocytic cups

Cells were fixed for 10 min in 3.7% formaldehyde solution andpermeabilized for 5 min in ice-cold acetone. Actin was stainedwith Alexa 568-labelled phalloidin (Molecular Probes; diluted1:200). We defined phagocytic cups as actin accumulationssurrounding the full circumference of the phagocytosed bead orbacterium. Actin nucleating Arp2/3 complex was stained with apolyclonal antibody recognizing the p41±Arc subunit (diluted1:100), WASp with antibody 3D8.H5 (diluted 1:50; Linder et al.,1999) recognizing an epitope between amino acids 102 and 201of WASp, human av-integrin receptor with monoclonal antibody(MAb) 1953Z-20 (Clone P3G8, Chemicon; diluted 1:100),human b1 integrin receptor with MAb 1951Z-20 (Clone P4G11,Chemicon; diluted 1:100), human b2 integrin receptor with MAb

Yersina invasion-induced phagocytosis is regulated by WASp 699

Q 2001 Blackwell Science Ltd, Cellular Microbiology, 3, 693±702

1962Z-20 (Clone P4H9, Chemicon; diluted 1: 100), humanaVb3 integrin receptor with MAb 1976Z-20 (Clone LM609,Chemicon; diluted 1:100), human aVb5 integrin receptor withMAb 1961Z-20 (Clone P1F6, Chemicon; diluted 1:100), humana5b1 integrin receptor with MAb 1999Z-20 (Clone HA6,Chemicon; diluted 1:100). Secondary antibodies were Alexa350, 488- and 568-labelled goat anti-mouse or goat anti-rabbit(Molecular Probes; diluted 1:200 each). Coverslips weremounted in Mowiol (Calbiochem) containing p-phenylenedia-mine (Sigma) as antifading reagent and sealed with nail polish.

Inhibitors

Anti-human b1 Integrin, MAb Ascites (Clone P4C10) waspurchased from Gibco-BRL; RGD and RGDS were pur-chased from Sigma and diluted in RPMI 1640 at theconcentrations indicated.

Quantification of internalized invasin latex beads

Macrophages cultured on coverslips were washed three timesand incubated with RPMI 1640 medium before addition ofbeads to obtain 50±100 attached beads per cell. Invasin-coated latex beads were centrifuged onto the cells (800 r.p.m.,3 min, 48C). After an incubation at 378C in a humidifiedatmosphere of 5% CO2 for the times indicated, cells werewashed with ice-cold 1% BSA-PBS to remove unbound beads.For differential staining of adherent and internalized invasin-coated latex beads, cells were incubated with rabbit anti-invasin antibody (diluted 1:1000) or goat anti-GST (diluted1:100) at 48C. Cells were washed three times with 1% BSA±PBS to remove unbound antibody and incubated with Alexa350-, 488- or 568-labelled donkey anti-goat or goat anti-rabbitantibodies (Molecular Probes; diluted 1:200 each) at 48C. Bythis way, internalised beads (blue colour due to 365/415fluorescence) were distinguished from extracellular beads(pink colour due to overlay of blue and red fluorescence).Fluorescence was evaluated by 32 sections through ventralparts of cells, using a confocal microscope.

Recombinant proteins

C3-transferase, Rho-binding domain of Rho-kinase, N17Rac, CRIB domain of the CDC42Hs/Rac target PAK,N17CDC42Hs and WASp-domain constructs VC and Awere generated and expressed as glutathione-S-transferase(GST) fusions. Proteins were dialyzed against microinjectionbuffer (50 mM Tris-HCLpH7.5/150 mM NaCl/5 mM MgCl2),concentrated in Centricon or Microcon (Amicon), shock-frozen, and stored at 2808C. Purity was tested using SDS±PAGE and Coomassie staining.

Microinjection of proteins

Cells for microinjection experiments were cultured for 5±8 days. Microinjection was performed by using transjector5246 (Eppendorf) and a compic inject micromanipulator (CellBiology Trading). Proteins were injected into the cytoplasm at0.5±2 mg ml21. Injected cells were identified by co-injected

FITC-Dextran (0.1 mg ml21; Molecular Probes). Controlinjections were performed with GST.

Acknowledgements

This work was supported by Deutsche Forschungsgemeinschaft(DFG) Grant Ae11 and SFB 413. The expert technical assistance

of Claudia Trasak and Barbara BoÈhlig is greatly appreciated. We

thank JuÈrgen Heesemann and Peter C. Weber for continuoussupport. The antibody against p41±Arc was a generous gift from

Dr Henry Higgs, Salk Institute, La Jolla, CA, USA; the antibody

against WASp was a generous gift from Dr David Nelson,

National Cancer Institute, NIH, Bethesda, USA.

References

Aderem, A., and Underhill, D.M. (1999) Mechanisms of phago-cytosis in macrophages. Annu Rev Immunol 17: 593±623.

Aepfelbacher, M., Essler, M., Huber, E., Czech, A., and Weber,

P.C. (1996) Rho is a negative regulator of human monocytespreading. J Immunol 157: 5070±5075.

Aepfelbacher, M., Zumbihl, R., Ruckdeschel, K., Jacobi, C.A.,Barz, C., and Heesemann, J. (1999) The tranquilizing injection

of Yersinia proteins: a pathogen's strategy to resist host

defense. Biol Chem 380: 795±802.

Allen, L.H., and Aderem, A. (1995) A role for MARCKS, the alpha

isozyme of protein kinase C and myosin I in zymosan

phagocytosis by macrophages. J Exp Med 182: 829±840.

Andor, A., TruÈ lzsch, K., Essler, M., Wiedemann, A., Roggen-

kamp, A., Heesemann, J., Aepfelbacher, M. (2001) YopE of

Yersinia, a GAP for RhoGTPases selectively modulates Rac-dependent actin structures in endothelial cells. Cell Microbiol 3:

301±310.

Autenrieth, I.B., and Firsching, R. (1996) Penetration of M cells and

destruction of Peyer's patches by Yersinia enterocolitica: an

ultrastructural and histological study. J Med Microbiol 44: 285±

294.

Barz, C., Abahji, T.N., Trulzsch, K., and Heesemann, J. (2000)

The Yersinia Ser/Thr protein kinase YpkA/YopO directlyinteracts with the small GTPases RhoA and Rac-1. FEBS

Lett 482: 139±143.

Bishop, A.L., and Hall, A. (2000) Rho GTPases and their effectorproteins. Biochem J 2: 241±255.

Black, D.S., and Bliska, J.B. (1997) Identification of p130Cas as asubstrate of Yersinia YopH (Yop51), a bacterial protein tyrosine

phosphatase that translocates into mammalian cells and

targets focal adhesions. EMBO J 16: 2730±2744.

Blystone, S.D., Graham, I.L., Lindberg, F.P., and Brown, E.J.

(1994) Integrin alpha v beta 3 differentially regulates adhesive

and phagocytic functions of the fibronectin receptor alpha 5beta 1. J Cell Biol 127: 1129±1137.

Bourdoulous, S., Orend, G., MacKenna, D.A., Pasqualini, R., and

Ruoslahti, E. (1998) Fibronectin matrix regulates activation ofRHO and CDC42 GTPases and cell cycle progression. J Cell

Biol 143: 267±276.

Caron, E., and Hall, A. (1998) Identification of two distinct

mechanisms of phagocytosis controlled by different Rho

GTPases. Science 282: 1717±1721.

Cornelis, G.R., and Wolf-Watz, H. (1997) The Yersinia Yop

virulon: a bacterial system for subverting eukaryotic cells. Mol

Microbiol 23: 861±867.

Cox, D., Chang, P., Zhang, Q., Reddy, P.G., Bokoch, G.M., and

Greenberg, S. (1997) Requirements for both Rac1 and Cdc42

700 A. Wiedemann et al.

Q 2001 Blackwell Science Ltd, Cellular Microbiology, 3, 693±702

in membrane ruffling and phagocytosis in leukocytes. J ExpMed 186: 1487±1494.

Daniels, R.H., and Bokoch, G.M. (1999) p21-activated protein

kinase: a crucial component of morphological signalling?Trends Biochem Sci 24: 350±355.

Dersch, P., and Isberg, R.R. (1999) A region of the Yersinia

pseudotuberculosis invasin protein enhances integrin-

mediated uptake into mammalian cells and promotes self-association. EMBO J 18: 1199±1213.

Dukuzumuremyi, J.M., Rosqvist, R., Hallberg, B., Akerstrom, B.,

Wolf-Watz, H., and Schesser, K. (2000) The Yersinia proteinkinase A is a host-factor inducible RhoA/Rac-binding virulence

factor. J Biol Chem 275: 35281±35290.

Essler, M., Amano, M., Kruse, H.J., Kaibuchi, K., Weber, P.C.,and Aepfelbacher, M. (1998) Thrombin inactivates myosin light

chain phosphatase via Rho and its target Rho kinase in human

endothelial cells. J Biol Chem 273: 21867±21874.

FaÈ llman, M., Andersson, K., Hakansson, S., Magnusson, K.E.,Stendahl, O., and Wolf-Watz, H. (1995) Yersinia pseudotuber-

culosis inhibits Fc receptor-mediated phagocytosis in J774

cells. Infect Immun 63: 3117±3124.

Finlay, B.B., and Cossart, P. (1997) Exploitation of mammalian host

cell functions by bacterial pathogens. Science 276: 718±725.

Giancotti, F.G. (2000) Complexity and specificity of integrinsignalling. Nat Cell Biol 2: 13±14.

Higgs, H.N., and Pollard, T.D. (1999) Regulation of actin

polymerization by Arp2/3 complex and WASp/Scar proteins.J Biol Chem 274: 3253±3254.

Hy, S., Rockow, S., Tang, J., Nishimura, R., Skolnik, E.Y.,

Chen, M. et al. (1997) Wiskott±Aldrich syndrome protein is

associated with the adapter protein Grb2 and the epidermalgrowth factor receptor in living cells. Mol Biol Cell 8: 1709±

1721.

Isberg, R.R., and Leong, J.M. (1990) Multiple b1 chain integrinsare receptors for invasin, a protein that promotes bacterial

penetration into mammalian cells. Cell 60: 861±871.

Kim, A.S., Kakalis, L.T., Abdul-Mana, N., Liu, G.A., andRosen, M.K. (2000) Autoinhibition and activation mechan-

isms of the Wiskott±Aldrich syndrome protein. Nature 404:

151±158.

Leong, J.M., Morrissey, P.E., Marra, A., and Isberg, R.R. (1995)An aspartate residue of the Yersinia pseudotuberculosis

invasin protein that is critical for integrin binding. EMBO J 14:

422±431.

Leverrier, Y., Lorenzi, R., Blundell, M.P., Brickell, P., Kinnon,

C., Ridley, A.J., and Thrasher, A.J. (2001) Cutting edge:

The Wiskott±Aldrich syndrome protein is required for

efficient phagocytosis of apoptotic cells. J Immunol 166:4831±4834.

Linder, S., Nelson, D., Weiss, M., and Aepfelbacher, M. (1999)

Wiskott±Aldrich syndrome protein regulates podosomes inprimary human macrophages. Proc Natl Acad Sci USA 96:

9648±9653.

Linder, S., Higgs, H., HuÈfner, K., Schwarz, K., Pannicke, U., andAepfelbacher, M. (2000) The polarization defect of Wiskott±

Aldrich syndrome macrophages is linked to dislocalisation of

the Arp2/3 complex. J Immunol 165: 221±225.

Lorenzi, R., Brickell, P.M., Katz, D.R., Kinnon, C., and Thrasher,

A.J. (2000) Wiskott±Aldrich syndrome protein is necessary

for efficient IgG-mediated phagocytosis. Blood 95: 2943±

2946.

Machesky, L.M., and Hall, A. (1997) Role of actin polymerization

and adhesion to extracellular matrix in Rac- and Rho-induced

cytoskeletal reorganization. J Cell Biol 138: 913±926.

Mainiero, F., Soriani, A., Strippoli, R., Jacobelli, J., Gismondi, A.,Piccoli, L. et al. (2000) RAC1/P38 MAPK signaling pathway

controls 1 integrin-induced Interleukin-8 production in human

natural killer cells. Immunity 12: 7±16.

Marra, A., and Isberg, R.R. (1997) Invasin-dependent and

invasin-independent pathways for translocation of Yersinia

pseudotuberculosis across the Peyer's patch intestinal epithe-lium. Infect Immun 65: 3412±3421.

Massol, P., Montcourrier, P., Guillemot, J.C., and Chavrier, P.

(1998) Fc receptor-mediated phagocytosis requires CDC42

and Rac1. EMBO J 17: 6219±6229.

May, R.C., Caron, E., Hall, A., and Machesky, L.M. (2000)

Involvement of the Arp2/3 complex in phagocytosis mediated

by FcgR or CR3. Nat Cell Biol 2: 246±248.

Nonoyama, S., and Ochs, H.D. (1998) Characterization of the

Wiskott±Aldrich syndrome protein and its role in the disease.Curr Opin Immunol 10: 407±412.

Ofek, L., Goldhar, J., Keisari, Y., and Sharon, N. (1995)Nonopsonic Phagocytosis of Microorganisms. Annu Rev

Microbiol 49: 239±276.

Pepe, J.C., Wachtel, M.R., Wagar, E., and Miller, V.L. (1995)

Pathogenesis of defined invasion mutants of Yersinia enter-

ocolitica in a BALB/c mouse model of infection. Infect Immun63: 4837±4848.

Persson, C., Carballeira, N., Wolf-Watz, H., and Fallman, M.(1997) The PTPase YopH inhibits uptake of Yersinia, tyrosine

phosphorylation of p130Cas and FAK, and the associated

accumulation of these proteins in peripheral focal adhesions.

EMBO J 16: 2307±2318.

Pommier, C.G., Inada, S., Fries, L.F., Takahashi, T., Frank,

M.M., and Brown, E.J. (1983) Plasma fibronectin enhancesphagocytosis of opsonized particles by human peripheral blood

monocytes. J Exp Med 157: 1844±1854.

Price, L.S., Leng, J., Schwartz, M.A., and Bokoch, G.M. (1998)

Activation of Rac and Cdc42 by integrins mediates cell

spreading. Mol Biol Cell 9: 1863±1871.

Ren, X.D., Kiosses, W.B., and Schwartz, M.A. (1999) Regulation

of the small GTP-binding protein Rho by cell adhesion and thecytoskeleton. EMBO J 18: 578±585.

Rivero-Lezcano, O.M., Marcilla, A., Sameshima, J.H., andRobbins, K.C. (1998) Wiskott±Aldrich syndrome protein

physically associates with Nck through Src homology 3

domains. Mol Cell Biol 15: 5725±5731.

Schulte, R., Zumbihl, R., Kampik, D., Fauconnier, A., and

Autenrieth, I.B. (1998) Wortmannin blocks Yersinia inva-

sin-triggered internalization, but not interleukin-8 produc-tion by epithelial cells. Med Microbiol Immunol 187: 53±

60.

Schulte, R., Grassl, G.A., Preger, S., Fessele, S., Jacobi, C.A.,

Schaller, M. et al. (2000) Yersinia enterocolitica invasin protein

triggers IL-8 production in epithelial cells via activation of Relp65-p65 homodimers. FASEB J 14: 1471±1484.

Silverstein, S.C. (1995) Phagocytosis of microbes: insights andprospects. Trends Cell Biol 5: 141±142.

Sumi, T., Matsumoto, K., Takai, Y., and Nakamura. T. (1999)Cofilin phosphorylation and actin cytoskeletal dynamics regu-

lated by rho- and Cdc42-activated LIM-kinase 2. J Cell Biol

147: 1519±1532.

Vidarsson, G., and van de Winkel, J.G.J. (1998) Fc receptor and

complement receptor-mediated phagocytosis in host defence.

Curr Opin Infect Dis 11: 271±278.

Virji, M. (1996) Microbial utilization of human signalling mole-cules. Microbiology 142: 3319±3336.

von Pawel-Rammingen, U., Telepnev, M.V., Schmidt, G.,

Yersina invasion-induced phagocytosis is regulated by WASp 701

Q 2001 Blackwell Science Ltd, Cellular Microbiology, 3, 693±702

Aktories, K., Wolf-Watz, H., and Rosqvist, R. (2000) GAPactivity of the Yersinia YopE cytotoxin specifically targets the

Rho pathway: a mechanism for disruption of actin microfila-

ment structure. Mol Microbiol 36: 737±748.

Wright, S.D., Craigmyle, L.S., and Silverstein, S.C. (1983)Fibronectin and serum amyloid P component stimulate C3b-

and C3bi-mediated phagocytosis in cultured human mono-

cytes. J Exp Med 158: 1338±1343.Yang, N., Higuchi, O., Ohashi, K., Nagata, K., Wada, A.,

Kangawa, K. et al. (1998) Cofilin phosphorylation by LIM-

kinase 1 and its role in Rac-mediated actin reorganization.Nature 393: 809±812.

Young, V.B., Miller, V.L., Falkow, S., and Schoolnik, G.K. (1990)

Sequence, localization and function of the invasin protein of

Yersinia enterocolitica. Mol Microbiol 4: 1119±1128.Zumbihl, R., Aepfelbacher, M., Andor, A., Jacobi, C.A., Ruck-

deschel, K., Rouot, B. et al. (1999) The cytotoxin YopT of

Yersinia enterocolitica induces modification and cellular redis-tribution of the small GTP-binding protein RhoA. J Biol Chem

274: 29289±29293.

702 A. Wiedemann et al.

Q 2001 Blackwell Science Ltd, Cellular Microbiology, 3, 693±702