International Journal of Medical Microbiology xxx (2013) xxx– xxx
Contents lists available at SciVerse ScienceDirect
International Journal of Medical Microbiology
jo u r n al hom epage: www.elsev ier .com/ locate / i jmm
ole for the fibrinogen-binding proteins Coagulase and Efb in the Staphylococcusureus–Candida interaction
arsten Fehrmanna,b, Kerstin Jurkc, Anne Bertlingb,c, Gabriela Seidela, Wolfgang Fegelera,eate E. Kehrelb,c, Georg Petersa,b, Karsten Beckera,b, Christine Heilmanna,b,∗
Institute of Medical Microbiology, University Hospital of Münster, D-48149 Münster, GermanyThe Interdisciplinary Center for Clinical Research (IZKF) Münster, University Hospital of Münster, D-48149 Münster, GermanyExperimental and Clinical Haemostasis, Department of Anaesthesiology and Intensive Care, University Hospital of Münster, D-48149 Münster, Germany
r t i c l e i n f o
rticle history:eceived 27 September 2012eceived in revised form 31 January 2013ccepted 3 February 2013
eywords:taphylococcus aureusandida sp.ixed species biofilms
hage display libraryoagulasefb
a b s t r a c t
Staphylococcus aureus and Candida species are increasingly coisolated from implant-associated polymi-crobial infections creating an incremental health care problem. Synergistic effects between both generaseem to facilitate the formation of mixed S. aureus–Candida biofilms, which is thought to play a critical rolein coinfections with these microorganisms. To identify and characterize S. aureus factors involved in theinteraction with Candida species, we affinity-panned an S. aureus phage display library against Candidabiofilms in the presence or absence of fibrinogen. Repeatedly isolated clones contained DNA fragmentsencoding portions of the S. aureus fibrinogen-binding proteins coagulase or Efb. The coagulase binds toprothrombin in a 1:1 ratio thereby inducing a conformational change and non-proteolytic activation ofprothrombin, which in turn cleaves fibrinogen to fibrin. Efb has been known to inhibit opsonization. Tostudy the role of coagulase and Efb in the S. aureus–Candida cross-kingdom interaction, we performedflow-cytometric phagocytosis assays. Preincubation with coagulase reduced the phagocytosis of Candida
yeasts by granulocytes significantly and dose-dependently. By using confocal laser scanning microscopy,we demonstrated that the coagulase mediated the formation of fibrin surrounding the candidal cells.Furthermore, the addition of Efb significantly protected the yeasts against phagocytosis by granulocytesin a dose-dependent and saturable fashion. In conclusion, the inhibition of phagocytosis of Candida cellsby coagulase and Efb via two distinct mechanisms suggests that S. aureus might be beneficial for Candidato persist as it helps Candida to circumvent the host immune system.
36
37
38
39
40
41
42
43
44
45
46
ntroduction
Due to the increasing use of medical devices in the past decades,mplant-associated infections, such as bloodstream infections (BIs),rosthetic endocarditis and osteomyelitis, have become majorauses of morbidity and mortality (von Eiff et al., 2005; Campocciat al., 2006). Approximately 250,000 cases of BIs associated withndwelling devices, such as intravascular catheters or orthope-ic implants, occur every year in the United States. A nationwideurveillance study revealed that Staphylococcus aureus and Candidaepresent the second and fourth most common pathogen isolated,
Please cite this article in press as: Fehrmann, C., et al., Role for the fibaureus–Candida interaction. Int. J. Med. Microbiol. (2013), http://dx.doi.org
espectively (Wisplinghoff et al., 2004). Generally, fungal implantnfections are considered as an increasing problem and frequentlynvolve pathogenic Candida species, in particular Candida albicans
∗ Corresponding author at: Institute of Medical Microbiology, University Hospitalf Münster, Domagkstr. 10, D-48149 Münster, Germany. Tel.: +49 251 8355357;ax: +49 251 8355350.
roduce, and host components. It has been estimated that biofilmsccount for over half of all infections (Costerton et al., 1999).olymicrobial biofilms, in which microorganisms interact in a syn-rgistic or inhibitory fashion, represent an underinvestigated andncreasing health care problem (Shirtliff et al., 2009). Scanninglectron microscopy (SEM) demonstrated a mixed-species biofilmonsisting of staphylococcal and candidal cells on the surface of anntracardial Hickman catheter, which were also isolated from blood
hen the patient developed septicemia (Costerton et al., 1985).ixed-species biofilms were also identified in infections of silicone
oice and orthopedic prostheses, biliary stents, endotracheal tubes,nd acrylic dentures (Costerton et al., 1999; Klotz et al., 2007).taphylococcal and candidal cells in a biofilm are more resistanto antibiotics and to antifungal agents, respectively, and protectedgainst the human immune system, which frequently causes theecessity to remove the medical device (Costerton et al., 1999;arriott and Noverr, 2009; Ramage et al., 2001, 2002). Moreover,ixed bacterial–fungal infections might exhibit properties that are
istinct from single-species infections (Wargo and Hogan, 2006).Very little is known about the interaction of staphylococci with
andida yeasts to date. Importantly, the combined effect of C. albi-ans and S. aureus results in synergism and increased mortality inice (Carlson, 1982). In S. aureus–Candida coinfections, the interac-
ions of the pathogens may lead to the formation of a mixed-speciesiofilm. Such interactions may be direct or mediated by bridgingolecules, such as the extracellular matrix and plasma proteins fib-
inogen (Fg) and fibronectin (Fn), because both pathogens expresshe respective surface-exposed binding proteins (Hostetter, 1994;aur and Klotz, 1997; Clarke and Foster, 2006). Moreover, interac-
ions leading to synergistic or inhibitory effects among both speciesight occur. To identify S. aureus factors involved in the interactionith Candida, we used the phage display technique and affinity-
elected phagemids that specifically bound to Candida. Isolatedhagemids contained portions of the Fg-binding proteins coagu-
ase or Efb. Phagocytosis assays revealed that the presence of both,oagulase and Efb, significantly and dose-dependently inhibited thehagocytosis of the clinically most important yeast species C. albi-ans and the closely related C. dubliniensis suggesting that S. aureusight be beneficial for Candida to persist as it helps the yeast to
vade the human immune system.
aterials and methods
acterial and candidal strains, phage display library, vector,edia, and reagents
S. aureus 8325-4 (Novick, 1963) and the clinical strain S. aureus074 that was isolated from a patient with native valve endo-arditis (Heilmann et al., 2002) was used to clone the coa andfb gene, respectively. Strains of C. albicans (MF6, MF7, MF14)nd C. dubliniensis (MZ44, MY29) were recovered from patientst the Charité in Berlin (Germany) and the University Hospital ofünster (Germany), respectively. Candida parapsilosis ATCC 22019
erved as a negative control. The previously described phage dis-lay library of S. aureus 4074 was used in the pannings (Heilmannt al., 2004). E. coli TG1 and helper phage R408 (Promega, Madison,I, USA) served to produce enriched phage stocks (Heilmann et al.,
004). The PCR-amplified coa or efb genes or N- or C-terminal por-ions of efb were cloned into the vector pQE30 Xa (Qiagen, Hilden,ermany). The E. coli strains K12 UT5600 (OmpT protease defi-ient) (Grodberg and Dunn, 1988) or M15 (pREP4) (Qiagen) served
Please cite this article in press as: Fehrmann, C., et al., Role for the fibaureus–Candida interaction. Int. J. Med. Microbiol. (2013), http://dx.doi.org
o purify His-rCoa- or His-rEfb-fusion proteins.Staphylococcal strains were grown on Tryptic Soy (TS) agar
lates or in TS broth (TSB) containing 0.25% glucose, when appro-riate. Candidal strains were cultivated in RPMI (Sigma Aldrich,
PRESSedical Microbiology xxx (2013) xxx– xxx
Munich, Germany) or YST medium (Sifin, Berlin, Germany) or on
Kimmig plates (Merck, Darmstadt, Germany). E. coli strains were
cultivated in Luria-Bertani (LB) medium (Difco), on LB agar plates,
which contained 50 �g/ml ampicillin and 25 �g/ml kanamycin
when appropriate (Sambrook et al., 1989). To maintain the F-
plasmid, E. coli TG1 was grown on M9-minimal agar (Sambrook
et al., 1989).
Bovine serum albumin (BSA) was purchased from Applichem
and human serum (off the clot) was from PAA, Cölbe, Germany.
Human proteins were obtained from Enzyme Research Lab-
oratories (Fg, prothrombin, �-thrombin) or Acris Antibodies
(Hiddenhausen, Germany) (C3b). Antibodies were purchased from
edium for 30 min at RT and washed again. In parallel experiments,he wells were then incubated with physiological concentrationsf Fg (2 mg/ml in RPMI) and washed again. Afterwards, 200 �l ofhe phagemid library in RPMI were added to the wells, and the
icrotiter plates were incubated for 4 h at RT. Washing and elutionf bound hybrid phages and the production of secondary phagetocks was carried out as described before (Heilmann et al., 2002,004). After three rounds of identical pannings, aliquots of the elu-tes were used to infect E. coli cells, resulting in the colonies thatere analyzed.
creening for the expression of the E-tag
Expression of the E-tag was detected by colony blot analysisssentially as described before (Heilmann et al., 2004; Jacobsson etl., 2003) except that the E-tag was detected by mouse anti-E-tagntibodies (Amersham Biosciences) and AP-conjugated anti-mousentibodies (Dako).
onstruction and purification of N-terminal 6xHis-rCoa andxHis-rEfb fusion proteins
The primers Coa-F/Coa-R or Efb-F/Efb-R were used to amplifyhe coa or efb gene, respectively. DNA fragments encodinghe N-terminal (rEfb-N) or C-terminal (rEfb-C) portion of Efbere amplified with the primers Efb-F/Efb-NR or Efb-CF/Efb-R,
espectively. The PCR-amplified fragments were cloned into theector pQE30Xa previously linearized by StuI/BamHI (rCoa) oramHI/KpnI (rEfb). Fusion proteins were purified under native con-itions using Ni-NTA affinity chromatography (Qiagen). The yieldf the fusion proteins per 100 ml culture volume as determined byhe Coomassie Plus (Bradford) protein assay (Pierce, Perbio Science,onn, Germany) were approximately: 750 �g for His-rCoa, 590 �g
or His-rEfb, 560 �g for His-rEfb-N, and 860 �g for His-rEfb-C.For use in the phagocytosis assays, His-rCoa was treated with
actor Xa protease to remove the N-terminal His-tag (Qiagen)esulting in a recombinant coagulase free of any vector-derivedmino acids (aa) at the N-terminus, which is crucial for its activityFriedrich et al., 2003). The activity of rCoa was verified by the cleav-ge of the chromogenic substrate Pefachrome® TH (Pentapharm,asel, Switzerland). The purified His-rEfb, His-rEfb-N, and His-rEfb-
were functionally analyzed for their capacities to bind to Fg or tohe human complement component 3b (C3b) in an ELISA adher-nce assay essentially as described before except that the wells of6-well microtiter plates were coated overnight with human Fg20 �g/ml), C3b (2.5 �g/ml), or 1% BSA in PBS at 4 ◦C (Hirschhausent al., 2010).
low-cytometric phagocytosis assay
All phagocytosis experiments were performed with the bloodonors giving informed consent according to human experi-entation guidelines. The phagocytosis assay was essentially
erformed according to Mollnes et al. (2002). Yeast cellsere grown on Kimmig-agar plates for 48 h, labeled withuorescein-isothiocyanate (FITC isomer I [100 �g/ml], Invitrogen)
n an appropriate staining buffer (10% DMSO, 0.3 �M Na2CO3,.3 �M NaHCO3) for 30 min at RT, washed, and adjusted to.6 × 106 cells/ml. Whole blood from a healthy adult volunteer100 �l containing 5.6 × 106 granulocytes/ml as determined by aell-counter system 900; Serono Baker Diagnostics, USA; 5:1 [v/v]n 3.18% Na-citrate) was coincubated with yeasts (1:1) for 2 h
Please cite this article in press as: Fehrmann, C., et al., Role for the fibaureus–Candida interaction. Int. J. Med. Microbiol. (2013), http://dx.doi.org
t 37 ◦C and 5% CO2. The cells were fixed with formaldehydend washed with PBS. External yeasts were quenched by 0.4%rypan-blue (Sigma) and erythrocytes were lysed. The cells wereesuspended in 500 �l PBS and analyzed on a FacsCALIBURTM (BD
PRESSedical Microbiology xxx (2013) xxx– xxx 3
Bioscience). Electronic gating was used to analyze 10,000 granu-
locytes in each sample. The FL1 photomultiplier (transmittance at
530 nm) was used to detect uptake of Candida-cells by granulocytes.
In parallel experiments, yeast cells were preincubated with pro-
thrombin (10 nM), Fg (1 mg/ml), and varying concentrations of
purified rCoa (0.5–5 nM) for 10 min at 37 ◦C. As a positive control,
yeast cells were preincubated with 0.1 U (1 nM) �-thrombin and Fg
(1 mg/ml). To ensure that observed effects are due to the activities
of the active rCoa•prothrombin complex or thrombin and not due
to secondary effects, 10 �g/ml argatroban (Argatra®; Mitsubishi
Pharma) was added in control assays either before or after the
preincubation step. In control experiments, preincubation was per-
formed with Factor Xa to exclude an influence of tracing amounts of
Factor Xa. To analyze the influence of rEfb on phagocytosis, different
concentrations of His-rEfb, His-rEfb-N, or His-rEfb-C (8 nM–4 �M)
were preincubated with anticoagulated blood for 30 min at 37 ◦C
and 5% CO2.
Scanning electron microscopy (SEM) and confocal laser scanning
microscopy (CSLM)
SEM from an ex vivo intravenous catheter coinfected with C.
albicans and S. aureus was performed as described before (Peters
et al., 1982). To visualize the formation of fibrin on the candidal
surface, C. dubliniensis MZ44 cells were FITC-labeled, incubated
with prothrombin (10 nM), Fg (0.25 mg/ml), the peptide GPRP
(1.25 mM), which inhibits fibrin crosslinking, and different concen-
trations of rCoa (0, 1.5, or 2.5 nM) or �-thrombin (1 nM) as a positive
control in 100 �l PBS for 10 min at 37 ◦C. Fibrin was detected by
a monoclonal mouse anti-human fibrin antibody (5 �g/ml) and a
PE-conjugated anti-mouse antibody (25 �g/ml). The probes were
analyzed on a Nikon Eclipse TE300 confocal laser scanning micro-
scope.
Statistical analysis
Statistical significance was analyzed by an unpaired Student’s
t-test. P values ≤0.05 were considered statistically significant andare indicated by asterisks: * (P ≤ 0.05), ** (P ≤ 0.005).
Results
C. albicans and S. aureus can be observed as coinfecting agents
in polymicrobial implant-associated infections: SEM demonstrated
that C. albicans forms microcolonies on intravenous catheters, to
which S. aureus can attach (Fig. 1).
Biofilm forming capacity of candidal strains
For the selection of a suitable candidal strain and conditions to
be used in the panning experiments, the biofilm forming capacities
of different candidal strains were analyzed by using a quantita-
tive biofilm assay adapted for candidal strains (Heilmann et al.,
1996a). The strains C. dubliniensis MZ44 and MY29 and C. albicans
MF7 formed the strongest biofilms, when they were grown in RPMI
with C. dubliniensis MZ44 showing the strongest biofilm forming
capacity (Fig. 2). Biofilm formation in YST was less pronounced
and almost absent in TSB with these strains. Biofilm formation of
C. albicans MF6 and MF14 was comparable in RPMI and YST and
rinogen-binding proteins Coagulase and Efb in the Staphylococcus/10.1016/j.ijmm.2013.02.011
also almost absent in TSB. C. parapsilosis ATCC 22019 that served as 289
a negative control did not form a considerable amount of biofilm 290
under any of these conditions. Because of its strong biofilm forming 291
capacity, C. dubliniensis MZ44 was selected for further experiments. 292
Fig. 1. Scanning electron micrograph of an early polymicrobial biofilm on an ex vivoia
P293
b294
295
a296
(297
d298
a299
p300
3301
A302
s303
2304
r305
r306
c307
FMRawiwie
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
ntravenous catheter infected with C. albicans, which forms microcolonies that serves binding foci for attaching S. aureus cells.
anning of the phage display library against C. dubliniensis MZ44iofilms
To identify domain(s) of S. aureus proteins specifically inter-cting with Candida, a previously described phage display libraryHeilmann et al., 2004) was affinity-panned against biofilms of C.ubliniensis MZ44 in a microtiter plate. After the third panning,n approximate 4500-fold enrichment of Candida affinity-selectedhages was observed. The enrichment after the third panning was8,000-fold, when the Candida biofilm was preincubated with Fg.s shown before, this enrichment is specific, when bacteria pos-ess a receptor for the ligand used in the panning (Heilmann et al.,
Please cite this article in press as: Fehrmann, C., et al., Role for the fibaureus–Candida interaction. Int. J. Med. Microbiol. (2013), http://dx.doi.org
002; Jacobsson and Frykberg, 1996). To identify potentially cor-ect clones, i.e. phagemid clones that carry an insert with an openeading frame (orf) correcting the reading frame so that the artifi-ial E-tag and thus gene VIII (encoding the major coat protein pVIII)
ig. 2. Quantitative assay of biofilm formation of C. dubliniensis strains (MZ44,Y29), C. albicans strains (MF6, MF7, MF14), and C. parapsilosis (ATCC 22019) in
PMI, YST, and TSB after 48 h. The biofilms were stained with safranin and read with VersaMax Microplate Reader at 490 nm. The strongest biofilm forming capacityas found, when C. dubliniensis MZ44 was grown in RPMI. Wells, to which ster-
le media lacking cells were added, served as controls; the values for these wellsere subtracted from the experimental readings. Each assay was performed at least
n triplicate and was performed at least three times on different days. Data arexpressed as mean ± SD (n = 3).
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
PRESSedical Microbiology xxx (2013) xxx– xxx
is expressed (Jacobsson et al., 2003; Heilmann et al., 2004), colony
blot analysis was performed (not shown) and E-tag positive clones
were further analyzed.
Analysis of affinity-selected clones by panning against C.
dubliniensis MZ44
Because of the strong enrichment of affinity-selected phages,plasmid DNA of E-tag positive colonies obtained after the thirdpanning against C. dubliniensis biofilms in the presence of Fg was
prepared and sequenced. Nucleotide sequence analysis of twenty
140-bp inserted DNA fragments (represented by clone C90) and one200-bp fragment (clone C101) revealed that they share identical
nucleotide sequences encoding the same protein domain. Com-
parison of the deduced amino acid (aa) sequences of clones C90(46 aa) and C101 (58 aa) with sequences of known proteins in the
Please cite this article in press as: Fehrmann, C., et al., Role for the fibrinogen-binding proteins Coagulase and Efb in the Staphylococcusaureus–Candida interaction. Int. J. Med. Microbiol. (2013), http://dx.doi.org/10.1016/j.ijmm.2013.02.011
ARTICLE IN PRESSG Model
IJMM 50684 1–9
C. Fehrmann et al. / International Journal of Medical Microbiology xxx (2013) xxx– xxx 5
Fig. 3. (A) Schematic map of the Fg-binding protein coagulase of strain 4074 showing the alignment with the polypeptide sequences expressed by the phagemid clonesC90 and C101 that were affinity-selected by panning against biofilms of Candida in the presence of Fg. SP, signal peptide; R, Fg-binding repeat domain. (B) Alignment of theamino acid sequences expressed by the phagemid clones C90 and C101 with the respective sequences from coagulases from strain 4074, COL, and 8325-4. Asterisks indicateidentical amino acids; colons indicate very similar amino acids; periods indicate somewhat similar amino acids; and arrows indicate the positions of the repeats.
Fig. 4. (A) Schematic map of the Fg-binding protein Efb of strain 4074 showing the alignment with the polypeptide sequences expressed by the phagemid clone C77 thatwas affinity selected by panning against biofilms of Candida in the presence of Fg. SP, signal peptide; R, Fg-binding repeat domain. The N-terminal (rEfb-N) and C-terminal(rEfb-C) portion of Efb that were expressed and purified as His-fusion proteins are indicated. (B) Alignment of the amino acid sequences expressed by the phagemid cloneC77 with the respective sequences from Efb from strain 4074 and 8325-4. Asterisks indicate identical amino acids and arrows indicate the positions of the repeats.
Table 1Panning against biofilms of C. dubliniensis with phages derived from clone C77 or C90 in a dilution series mixed with a constant amount of phages derived from the phagemidvector pG8SAET. Given are the relative numbers (%) calculated from 16 clones analyzed from each condition.
Fig. 5. Flow-cytometric phagocytosis assay of C. dubliniensis MZ44 by granulocytesin whole blood, which was set to 100% phagocytosis (MZ44) and after preincuba-tion with prothrombin (10 nM), Fg (1 mg/ml), and rCoa (0.5–5 nM). As a positivecontrol, preincubation was performed with �-thrombin (1 nM). Preincubation withrCoa dose-dependently inhibited the phagocytosis of C. dubliniensis by granulocytes,which reached 70% and was statistically significant with 5 nM rCoa (white bars). Thiseffect could be inhibited by preincubation with the thrombin inhibitor argatroban(black bars), which blocks the active center of the activated rCoa•prothrombin com-plex or thrombin, but not when argatroban was added after preincubation (dashedbam
fi367
c368
I369
370
i371
p372
t373
F374
w375
t376
a377
t378
c379
p380
r381
d382
s383
r384
m385
(386
c387
w388
C389
s390
8391
b392
D393
394
d395
c396
o397
d398
c399
n400
t401
2402
2403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
ar) indicating that the observed effect indeed depends on thrombin activity. Resultsre shown as the mean of three independent experiments. Statistical significance isarked by asterisks: * (P ≤ 0.05), ** (P ≤ 0.005).
To elucidate the role that Efb plays in the S. aureus–Candidanteraction, we analyzed the effect of purified rEfb, the N-terminalortion of Efb (rEfb-N), or the C-terminal portion of Efb (rEfb-C) onhe phagocytosis of Candida cells by granulocytes in whole blood.or this, we expressed the nucleotide sequences encoding thehole rEfb, rEfb-N, or rEfb-C in E. coli (Fig. 4), purified the respec-
ive gene products, and performed flow-cytometric phagocytosisssays. Previously, the activities of these proteins (rEfb: bindingo Fg and C3b; rEfb-N: binding to Fg; rEfb-C: binding to C3b) wereonfirmed in ELISA adherence assays (not shown). Flow-cytometrichagocytosis assays revealed that rEfb and to a lesser extentEfb-C inhibited the phagocytosis of C. albicans MF6 in a dose-ependent and saturable fashion, which was statistically significanttarting with 400 nM rEfb (88% ± 6.6% phagocytosis) and 500 nMEfb-C (83% ± 4.8% phagocytosis) (Fig. 7A). Inhibition was maxi-al with 800 nM rEfb (62.5% ± 8% phagocytosis) and 2 �M rEfb-C
82% ± 2.5% phagocytosis). rEfb-N only slightly inhibited phago-ytosis (91.4% ± 3.7% phagocytosis with 500 nM), which howeveras not statistically significant. Similar results were obtained for
. dubliniensis MZ44 (Fig. 7B). Maximal inhibition of phagocyto-is was observed with 800 nM rEfb (60% ± 9% phagocytosis) and00 nM rEfb-C (83.5% ± 5.4% phagocytosis). As a negative control,uffer without purified protein was added, which had no effect.
iscussion
In the past 15 years, substantial progress has been made inefining the molecular mechanisms that are involved in staphylo-occal biofilm formation (Heilmann, 2011; Götz, 2002). Adherencef staphylococci to polymeric or biotic surfaces can occur eitherirectly or mediated by bridging molecules, such as the extra-ellular matrix and plasma proteins Fg and Fn, and involves a
Please cite this article in press as: Fehrmann, C., et al., Role for the fibaureus–Candida interaction. Int. J. Med. Microbiol. (2013), http://dx.doi.org
umber of different surface-anchored and surface-associated pro-eins (Chavakis et al., 2005; Clarke and Foster, 2006; Heilmann,011; Götz, 2002; Heilmann et al., 1997, 2005; Wann et al.,000). Subsequently, the bacteria proliferate and accumulate into a
PRESSedical Microbiology xxx (2013) xxx– xxx
multilayered, three-dimensional biofilm structure, which involves
the production of the polysaccharide intercellular adhesin (PIA)
that is an N-acetylglucosaminoglycan synthesized by the gene
products of the icaADBC operon (Cramton et al., 1999; Heilmann
et al., 1996a, b; Mack et al., 1996) or surface proteins, such as Aap
(Rohde et al., 2005), SasG (Corrigan et al., 2007), FnBPs (O’Neill et al.,
2008), and SasC (Schroeder et al., 2009). In Candida biofilm forma-
tion, adherence of yeast cells to a polymeric surface is followed by
germ tube formation finally resulting in a biofilm that consists of a
mixture of yeast and hyphal forms embedded within exopolymeric
material (Ramage et al., 2001; Kuhn et al., 2002). Genes encoding
surface proteins of C. albicans seem to be involved in biofilm for-
mation, i.e. genes of the ALS (agglutinin-like sequence) gene family
(O’Connor et al., 2005; Sheppard et al., 2004) as well as the adhesin
genes HWP1 (Staab et al., 2004) and EAP1 (Li and Palecek, 2003).
So far, there has been very little known about the interaction ofstaphylococci with Candida yeasts.
To establish an infection, a pathogen first needs to gain accessto the host usually by colonizing host tissues or medical devices
and forming a biofilm. Besides adherence and colonization, manypathogens developed strategies to evade the host immune sys-
tem, especially opsonophagocytosis and the complement system,
frequently resulting in persistent infection. One of the critical com-
ponents of the complement cascade is the complement protein
C3. This protein not only plays a crucial role as a precursor of an
opsonin, but is also the common link between the three pathways
(classical, alternative, and mannose-binding lectin) of complement
activation. The activation of the complement pathway leads to
the deposition of the C3-cleavage product C3b on the surface of
bacteria, resulting in opsonophagocytic killing of the bacteria. S.
aureus and many other human pathogens produce immunomod-
ulatory molecules that interfere with components of the human
immune system (Rooijakkers et al., 2005). In settings of mixed-
species biofilms, adhesive and immunomodulatory interactions
between different species, e.g. staphylococci and Candida, are very
likely.
To our knowledge, we here describe for the first time the iden-
tification of S. aureus proteins that interact with Candida yeasts
by using the phage display technique. In this study, we were
interested in identifying S. aureus surface factors involved in the
interaction with Candida. In coadherence and coaggregation assays,
we found pronounced adhesive interactions especially between
the clinical strain S. aureus 4074 that was used to construct the
phage display library and C. dubliniensis MZ44 that was used in
the pannings (not shown). However, we did not isolate phagemids
that contained portions of S. aureus surface-anchored or surface-
associated proteins potentially involved in this interaction. Instead,
we affinity-selected phagemids that contained portions of the
secreted and extracellular Fg-binding proteins Efb or coagulase
with Candida biofilms preincubated with Fg. Most likely, Efb and
the coagulase do not confer binding of S. aureus to Candida. Because
these proteins are thought to interfere with the host immune sys-
tem, we analyzed their potential impact on opsonophagocytosis of
Candida cells, which might be of great importance in settings of S.
aureus–Candida coinfections.
Efb is a constitutively and in vivo secreted protein that can
also interfere with platelet aggregation and delays would heal-
ing (Palma et al., 1996; Shannon and Flock, 2004). Recent studies
demonstrated that Efb not only binds Fg via its N-terminal domain,
but can simultaneously also bind the complement component C3b
via its C-terminal domain (Lee et al., 2004b). Upon binding of Efb
to the �-chain of C3, all three pathways of complement activation
rinogen-binding proteins Coagulase and Efb in the Staphylococcus/10.1016/j.ijmm.2013.02.011
seem to be inhibited (Lee et al., 2004a). Additionally, Efb can inhibit 466
complement-mediated opsonophagocytosis (Lee et al., 2004a). We 467
found that rEfb and to a lesser extent also rEfb-C significantly and 468
dose-dependently inhibited phagocytosis of Candida probably by 469
C. Fehrmann et al. / International Journal of Medical Microbiology xxx (2013) xxx– xxx 7
Fig. 6. Confocal laser scanning micrographs of coagulase-mediated fibrin formation (red) surrounding the cells of C. dubliniensis MZ44 (green). Fibrin formation surroundinga ationsf ion wQ3r
b470
m471
s472
t473
t474
C475
I476
o477
p478
479
1480
p481
c482
(483
s484
c485
o486
t487
b488
o489
s490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
nd covering the yeast cells (merge: yellow) was visible at low coagulase concentrormation of fibrin fibers (D). In the negative control without rCoa, no fibrin formateferred to the web version of the article.)
locking C3b deposition to the Candida surface suggesting that in aixed S. aureus–Candida biofilm, rEfb may facilitate both, the per-
istence of the S. aureus as well as the Candida infection. It seemshat for its full function, the whole Efb is required that binds N-erminally to Candida probably via Fg as bridging molecule and-terminally to C3 thereby preventing the cleavage of C3 to C3b.
ndeed, in flow-cytometric adherence assays, we found bindingf rEfb and rEfb-N, but not rEfb-C to the candidal surface in theresence of Fg (not shown).
Coagulase binds to prothrombin in a stoichiometric ratio of:1 leading to a conformational change and subsequent non-roteolytic activation of prothrombin (Hendrix et al., 1983). Theoagulase•prothrombin complex specifically cleaves Fg to fibrinPanizzi et al., 2004; Friedrich et al., 2003). Because we affinity-elected the Fg-binding portion of coagulase in an approach usingandidal biofilms preincubated with Fg, we assume that bindingccurs via Candida-bound Fg as bridging molecule. In that case,
Please cite this article in press as: Fehrmann, C., et al., Role for the fibaureus–Candida interaction. Int. J. Med. Microbiol. (2013), http://dx.doi.org
he N-terminus, which is responsible for activation of prothrom-in, might be available for prothrombin activation at the surfacef Candida (Friedrich et al., 2003). Accordingly, we could demon-trate the formation of fibrin surrounding Candida cells by CLSM
(1.5 or 2.5 nM) (B, C). Similarly, the positive control �-thrombin (1 nM) led to theas visible (A). (For interpretation of the references to color in the text, the reader is
(Fig. 6). This fibrin shield may protect the yeast against phago-
cytosis. Indeed, phagocytosis assays revealed that the presence of
rCoa reduced the phagocytosis of Candida significantly and dose-
dependently. Besides Fg, thrombin cleaves other substrates that
are present in the whole blood setting, such as various coagula-tion factors, protein C, and protease-activated receptors (PAR)-1, -3
and -4 (Gallwitz et al., 2012). Therefore, we cannot completely rule
out the possibility that additionally, some indirect effects may hap-
pen. However, as the concentration of Fg in this scenario is much
higher than that of any of the other factors mentioned, the contribu-
tion of any of such indirect effects might be very low or negligible.
Thus, we here confirmed an immunomodulatory role for coagulase,
i.e. the inhibition of phagocytosis. Earlier investigations suggested
that the coagulase does not have a function in the initiation of an
infection, because in a rat endocarditis model, a coagulase-negativemutant was not attenuated 12 h after bacterial challenge (Moreillon
et al., 1995). However, recently the coagulase was found to be a
rinogen-binding proteins Coagulase and Efb in the Staphylococcus/10.1016/j.ijmm.2013.02.011
virulence factor in three different mouse models for staphylococ- 508
cal disease: survival in blood, lethal bacteremia, and renal abscess 509
formation (Cheng et al., 2010). Furthermore, in a murine model of 510
S. aureus hematogenous pulmonary infection, significantly more 511
Fig. 7. Flow-cytometric analysis of the influence of rEfb, rEfb-N, or rEfb-C on thephagocytosis of C. albicans MF6 (A) or C. dubliniensis MZ44 (B) by granulocytes inwhole blood, which was set to 100% phagocytosis (white bars). rEfb and to a lesserextent rEfb-C dose-dependently inhibited the phagocytosis of C. albicans MF6, whichwas statistically significant starting with 400 nM rEfb (black bars) and 500 nM rEfb-C(diagonally dashed bars). rEfb-N (vertically dashed bars) slightly inhibited phagocy-tosis, which however was not statistically significant. Similar results were obtainedfor C. dubliniensis MZ44. As a negative control, buffer without purified protein wasadded, which had no effect (white bars). Results are shown as the mean of three inde-p(
w512
r513
t514
t515
r516
p517
i518
t519
(520
i521
o522
523
p524
i525
t526
p527
b528
s529
c530
t531
h532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
Q2 597
598
599
600
601
602
603
604
endent experiments. Statistical significance is marked by asterisks: * (P ≤ 0.05), **P ≤ 0.005).
ild-type cells compared to coagulase-negative mutant cells wereecovered from infected lungs 7 days after infection, suggestinghat the coagulase might promote bacterial proliferation in theissue during later stages of infection (Sawai et al., 1997). Ouresults indicate that this might be due to its inhibitory effect onhagocytosis. In agreement, most recent findings have shown that
n an in vitro infection model, S. aureus is protected from neu-rophils via a coagulase-dependent barrier termed pseudocapsuleGuggenberger et al., 2012). Significantly, our results suggest thatn S. aureus–Candida coinfections, also Candida can take advantagef the phagocytosis-inhibiting effect of the coagulase.
In this report, we demonstrate that secreted factors of aathogen may be utilized by another species to circumvent the host
mmune system. Efb and coagulase are virulence factors of S. aureushat may facilitate not only persistent S. aureus infections, but alsoersistent Candida infections in settings of polymicrobial infectionsy interfering with different lines of defense of the host immune
Please cite this article in press as: Fehrmann, C., et al., Role for the fibaureus–Candida interaction. Int. J. Med. Microbiol. (2013), http://dx.doi.org
ystem. In conclusion, the inhibition of phagocytosis of Candidaells by coagulase and Efb via two distinct mechanisms suggestshat S. aureus has a synergistic effect on Candida persistence as itelps the yeast to circumvent the host immune system.
PRESSedical Microbiology xxx (2013) xxx– xxx
Acknowledgments
We thank K. Schroeder, E. Kruse, and S. Weber for excellent
technical assistance. M. Ruhnke is acknowledged for providing the
Candida albicans clinical isolates and C. Neumann is acknowledged
for help in the construction of the Escherichia coli clone expressing
coa. This work was supported by a grant to C.H. and K.B. by the Inter-
disciplinary Clinical Research Center (IZKF) (Hei2/042/04) and the
Deutsche Forschungsgemeinschaft (DFG) (HE 3546/3-1) and par-
tially by a grant to C.H., B.E.K., and G.P. by the DFG, Collaborative
Research Center 293, Project A6. This study is part of the PhD thesis
of C.F.
References
Campoccia, D., Montanaro, L., Arciola, C.R., 2006. The significance of infection
related to orthopedic devices and issues of antibiotic resistance. Biomaterials27, 2331–2339.
Carlson, E., 1982. Synergistic effect of Candida albicans and Staphylococcus aureus on
mouse mortality. Infect. Immun. 38, 921–924.
Chavakis, T., Wiechmann, K., Preissner, K.T., Herrmann, M., 2005. Staphylococcusaureus interactions with the endothelium: the role of bacterial “secretable
expanded repertoire adhesive molecules” (SERAM) in disturbing host defense
systems. Thromb. Haemost. 94, 278–285.
Cheng, A.G., McAdow, M., Kim, H.K., Bae, T., Missiakas, D.M., Schneewind, O., 2010.Contribution of coagulases towards Staphylococcus aureus disease and protective
eilmann, C., Niemann, S., Sinha, B., Herrmann, M., Kehrel, B.E., Peters, G., 2004.Staphylococcus aureus fibronectin-binding protein (FnBP)-mediated adherenceto platelets, and aggregation of platelets induced by FnBPA but not by FnBPB. J.Infect. Dis. 190, 321–329.
eilmann, C., Schweitzer, O., Gerke, C., Vanittanakom, N., Mack, D., Götz, F., 1996b.Molecular basis of intercellular adhesion in the biofilm-forming Staphylococcusepidermidis. Mol. Microbiol. 20, 1083–1091.
endrix, H., Lindhout, T., Mertens, K., Engels, W., Hemker, H.C., 1983. Activation ofhuman prothrombin by stoichiometric levels of staphylocoagulase. J. Biol. Chem.258, 3637–3644.
irschhausen, N., Schlesier, T., Schmidt, M.A., Götz, F., Peters, G., Heilmann, C., 2010.A novel staphylococcal internalization mechanism involves the major autolysinAtl and heat shock cognate protein Hsc70 as host cell receptor. Cell. Microbiol.12, 1746–1764.
ostetter, M.K., 1994. Adhesins and ligands involved in the interaction of Candidaspp. with epithelial and endothelial surfaces. Clin. Microbiol. Rev. 7, 29–42.
ube, B., 2004. From commensal to pathogen: stage- and tissue-specific geneexpression of Candida albicans. Curr. Opin. Microbiol. 7, 336–341.
acobsson, K., Frykberg, L., 1996. Phage display shot-gun cloning of ligand-bindingdomains of prokaryotic receptors approaches 100% correct clones. Biotech-niques 20, 1070–1076, 1078, 1080–1071.
acobsson, K., Rosander, A., Bjerketorp, J., Frykberg, L., 2003. Shotgun phage display –selection for bacterial receptins or other exported proteins. Biol. Proced. Online5, 123–135.
uhn, D.M., Chandra, J., Mukherjee, P.K., Ghannoum, M.A., 2002. Comparison ofbiofilms formed by Candida albicans and Candida parapsilosis on bioprostheticsurfaces. Infect. Immun. 70, 878–888.
ee, L.Y., Höök, M., Haviland, D., Wetsel, R.A., Yonter, E.O., Syribeys, P., Vernachio, J.,Brown, E.L., 2004a. Inhibition of complement activation by a secreted Staphylo-coccus aureus protein. J. Infect. Dis. 190, 571–579.
ee, L.Y., Liang, X., Höök, M., Brown, E.L., 2004b. Identification and characterizationof the C3 binding domain of the Staphylococcus aureus extracellular fibrinogen-binding protein (Efb). J. Biol. Chem. 279, 50710–50716.
i, F., Palecek, S.P., 2003. EAP1, a Candida albicans gene involved in binding humanepithelial cells. Eukaryot. Cell. 2, 1266–1273.
ack, D., Fischer, W., Krokotsch, A., Leopold, K., Hartmann, R., Egge, H., Laufs, R.,1996. The intercellular adhesin involved in biofilm accumulation of Staphylo-coccus epidermidis is a linear beta-1,6-linked glucosaminoglycan: purificationand structural analysis. J. Bacteriol. 178, 175–183.
ollnes, T.E., Brekke, O.L., Fung, M., Fure, H., Christiansen, D., Bergseth, G., Videm, V.,Lappegard, K.T., Kohl, J., Lambris, J.D., 2002. Essential role of the C5a receptor inE coli-induced oxidative burst and phagocytosis revealed by a novel lepirudin-based human whole blood model of inflammation. Blood 100, 1869–1877.
oreillon, P., Entenza, J.M., Francioli, P., McDevitt, D., Foster, T.J., Francois, P., Vau-daux, P., 1995. Role of Staphylococcus aureus coagulase and clumping factor inpathogenesis of experimental endocarditis. Infect. Immun. 63, 4738–4743.
ovick, R.P., 1963. Properties of a cryptic high-frequency transducing phage inStaphylococcus aureus. Virology 33, 155–166.
’Connor, L., Lahiff, S., Casey, F., Glennon, M., Cormican, M., Maher, M., 2005. Quan-tification of ALS1 gene expression in Candida albicans biofilms by RT-PCR usinghybridisation probes on the LightCycler. Mol. Cell. Probes 19, 153–162.
’Neill, E., Pozzi, C., Houston, P., Humphreys, H., Robinson, D.A., Loughman, A., Foster,T.J., O’Gara, J.P., 2008. A novel Staphylococcus aureus biofilm phenotype medi-ated by the fibronectin-binding proteins, FnBPA and FnBPB. J. Bacteriol. 190,3835–3850.
alma, M., Nozohoor, S., Schennings, T., Heimdahl, A., Flock, J.I., 1996. Lack ofthe extracellular 19-kilodalton fibrinogen-binding protein from Staphylococcusaureus decreases virulence in experimental wound infection. Infect. Immun. 64,5284–5289.
Please cite this article in press as: Fehrmann, C., et al., Role for the fibaureus–Candida interaction. Int. J. Med. Microbiol. (2013), http://dx.doi.org
anizzi, P., Friedrich, R., Fuentes-Prior, P., Bode, W., Bock, P.E., 2004. The staphylo-coagulase family of zymogen activator and adhesion proteins. Cell. Mol. Life Sci.61, 2793–2798.
eters, G., Locci, R., Pulverer, G., 1982. Adherence and growth of coagulase-negativestaphylococci on surfaces of intravenous catheters. J. Infect. Dis. 146, 479–482.
Houston, A., Coffman, S., Hollis, R.J., 1999. International surveillance of blood
stream infections due to Candida species in the European SENTRY Program:
species distribution and antifungal susceptibility including the investigational
triazole and echinocandin agents, SENTRY Participant Group (Europe). Diagn.
Microbiol. Infect. Dis. 35, 19–25.
Pittet, D., Li, N., Wenzel, R.P., 1993. Association of secondary and polymicrobial
nosocomial bloodstream infections with higher mortality. Eur. J. Clin. Microbiol.Infect. Dis. 12, 813–819.
Ramage, G., Vande Walle, K., Wickes, B.L., Lopez-Ribot, J.L., 2001. Biofilm formationby Candida dubliniensis. J. Clin. Microbiol. 39, 3234–3240.
Ramage, G., VandeWalle, K., Bachmann, S.P., Wickes, B.L., Lopez-Ribot, J.L., 2002.In vitro pharmacodynamic properties of three antifungal agents against pre-
formed Candida albicans biofilms determined by time-kill studies. Antimicrob.Agents Chemother. 46, 3634–3636.
Rohde, H., Burdelski, C., Bartscht, K., Hussain, M., Buck, F., Horstkotte, M.A., Knobloch,J.K., Heilmann, C., Herrmann, M., Mack, D., 2005. Induction of Staphylococcus
epidermidis biofilm formation via proteolytic processing of the accumulation-associated protein by staphylococcal and host proteases. Mol. Microbiol. 55,
1883–1895.
Rooijakkers, S.H., van Kessel, K.P., van Strijp, J.A., 2005. Staphylococcal innateimmune evasion. Trends Microbiol. 13, 596–601.
Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A Laboratory Man-ual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York.
Sawai, T., Tomono, K., Yanagihara, K., Yamamoto, Y., Kaku, M., Hirakata, Y., Koga, H.,
Tashiro, T., Kohno, S., 1997. Role of coagulase in a murine model of hematogenous
pulmonary infection induced by intravenous injection of Staphylococcus aureusenmeshed in agar beads. Infect. Immun. 65, 466–471.
