Yersinia enterocolitica YadA Mediates
Complement Evasion by Recruitment and
Inactivation of C3 Products
This information is current as
of June 18, 2017.
Magnus K. H. Schindler, Monika S. Schütz, Melanie C.
Mühlenkamp, Suzan H. M. Rooijakkers, Teresia Hallström,
Peter F. Zipfel and Ingo B. Autenrieth
J Immunol published online 15 October 2012
http://www.jimmunol.org/content/early/2012/10/14/jimmun
ol.1201383
Permissions
Email Alerts
Information about subscribing to The Journal of Immunology is online at:
http://jimmunol.org/subscription
Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
Receive free email-alerts when new articles cite this article. Sign up at:
http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by
The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 2012 by The American Association of
Immunologists, Inc. All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
Subscription
Published October 15, 2012, doi:10.4049/jimmunol.1201383
The Journal of Immunology
Yersinia enterocolitica YadA Mediates Complement Evasion
by Recruitment and Inactivation of C3 Products
Magnus K. H. Schindler,*,1 Monika S. Schütz,*,1 Melanie C. Mühlenkamp,*
Suzan H. M. Rooijakkers,† Teresia Hallström,‡ Peter F. Zipfel,‡,x and Ingo B. Autenrieth*
Y
ersinia enterocolitica is a facultative anaerobic pleomorphic gram-negative rod that belongs to the family Enterobacteriaceae. Infections are caused by ingestion of contaminated food or drinking water and can cause severe diarrhea,
enterocolitis, and mesenteric lymphadenitis (1, 2). The pathogenicity of Y. enterocolitica is associated with the presence of a 70-kb
virulence plasmid (pYV) that encodes a type three secretion system, translocated effector proteins, and the trimeric autotransporter Yersinia adhesin A (YadA) (3).
YadA has manifold functions associated with the pathogenicity
of Y. enterocolitica. YadA mediates adhesion to host cells, promotes autoaggregation, and protects from complement-mediated
killing (4). Y. enterocolitica virulence in vivo is associated with the
presence of YadA, and it is known that the trimeric stability of
YadA correlates with pathogenicity in a mouse model of infection
(5). Being the prototype of trimeric autotransporter adhesins,
*Institute for Medical Microbiology and Hygiene, University Hospital Tübingen, 72076
Tübingen, Germany; †Medical Microbiology, University Medical Center Utrecht, 3584
CX Utrecht, The Netherlands; ‡Leibniz Institute for Natural Product Research and
Infection Biology-Hans Knöll Institute, 07745 Jena, Germany; and xFaculty of Biology
and Pharmacy, Friedrich Schiller University, 07743 Jena, Germany
1
M.K.H.S. and M.S.S. contributed equally to this work.
Received for publication May 18, 2012. Accepted for publication September 10,
2012.
M.S.S., I.B.A., and M.K.H.S. were supported by grants from the Deutsche Forschungsgemeinschaft within the collaborative research center Sonderforschungsbereich 766 “The bacterial cell envelope.” S.H.M.R. was supported by a Netherlands
Organisation for Scientific Research-Vidi grant.
Address correspondence and reprint requests to Dr. Monika S. Schütz, Institute
for Medical Microbiology and Hygiene, University Hospital Tübingen, ElfriedeAulhornstrasse 6, Tübingen D-72076, Germany. E-mail address: monika.schuetz@med.
uni-tuebingen.de
Abbreviations used in this article: AP, alternative pathway; CP, classical pathway;
HIS, heat-inactivated serum; LP, lectin pathway; NHS, normal human serum; TCC,
terminal complement complex; YadA, Yersinia adhesin A.
Copyright Ó 2012 by The American Association of Immunologists, Inc. 0022-1767/12/$16.00
www.jimmunol.org/cgi/doi/10.4049/jimmunol.1201383
YadA consists of three 50-kDa monomers that trimerize upon insertion into the bacterial outer membrane (6). The exchange of
a highly conserved glycine within the C-terminal b-barrel membrane anchor domain by amino acids with larger side chain size
(such as threonine, asparagine, histidine) reduces YadA trimer stability and eventually impaired passenger domain translocation.
YadA mutants that carried the small amino acids Alanine or Serine
instead of Glycine at position 389 resembled wild type expression
levels and phenotype except that they were sensitive to complementdependent killing and displayed reduced trimer stability in Western
blot (5, 7).
Complement resistance is a common mechanism of human and
animal pathogens to establish an infection in an immune competent
host. Several pathogens have evolved a plethora of mechanisms
that enable their evasion from complement-mediated killing (8–
11). The complement system is an important arm of the innate
immune system. It consists of .30 soluble and membrane-bound
proteins that initiate a cascade of catalytic cleavage reactions upon
contact with foreign surfaces and structures. Three pathways
of complement activation are distinguished: the classical pathway (CP), the alternative pathway (AP), and the lectin pathway
(LP). All three pathways are activated by recognition of non–self
structures on target surface (e.g., bacterial sugars) and result in the
formation of C3 convertases, which catalyze an important step in
the complement cascade: the cleavage of C3 into C3b. The C3
protein contains a reactive thioester group that is hidden within
the C3 molecule (12). Upon cleavage of C3, the thioester group
becomes exposed and reacts with the target surface to deposit
covalently bound C3b. Most of the generated C3b never attaches
to the bacterial surface because its thioester reacts with water
forming fluid phase C3b (C3bH2O), which is rapidly inactivated
by host regulators.
Surface-bound C3b molecules initiate a powerful amplification
loop to deposit more C3b on the surface. Bacteria covered with
C3b are effectively taken up by phagocytes, because C3b and its
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
Yersinia adhesin A (YadA) is a major virulence factor of Yersinia enterocolitica. YadA mediates host cell binding and autoaggregation and protects the pathogen from killing by the complement system. Previous studies demonstrated that YadA is the most
important single factor mediating serum resistance of Y. enterocolitica, presumably by binding C4b binding protein (C4BP) and
factor H, which are both complement inhibitors. Factor H acts as a cofactor for factor I-mediated cleavage of C3b into the inactive
form iC3b and thus prevents formation of inflammatory effector compounds and the terminal complement complex. In this study,
we challenged the current direct binding model of factor H to YadA and show that Y. enterocolitica YadA recruits C3b and iC3b
directly, without the need of an active complement cascade or additional serum factors. Enhanced binding of C3b does not
decrease survival of YadA-expressing Yersiniae because C3b becomes readily inactivated by factor H and factor I. Binding of
factor H to YadA is greatly reduced in the absence of C3. Experiments using Yersinia lacking YadA or expressing YadA with
reduced trimeric stability clearly demonstrate that both the presence and full trimeric stability of YadA are essential for
complement resistance. A novel mechanism of factor H binding is presented in which YadA exploits recruitment of C3b or
iC3b to attract large amounts of factor H. As a consequence, formation of the terminal complement complex is limited and
bacterial survival is enhanced. These findings add a new aspect of how Y. enterocolitica effectively evades the host complement
system. The Journal of Immunology, 2012, 189: 000–000.
2
YadA-MEDIATED COMPLEMENT EVASION
purchased from Complement Technology. The sera were aliquoted and
stored at 280˚C. Treatment at 56˚C for 30 min results in inactivation of the
complement cascade. The efficient depletion of DC6 and DC3 sera was
verified before by discrete application and quantification and testing their
ability to kill bacteria. Heat-treated sera were used in several experiments
and are termed heat-inactivated serum (HIS), HIDC3S, or HIDC6S
throughout the article. Purified proteins C3, C3b, and iC3b were purchased
from Complement Technology. As a result of unspecific background
staining of Yersinia, factor H (goat–anti-human; Complement Technology), and C5b-9 (rabbit–anti-human; Complement Technology), these Abs
were preabsorbed before usage. Therefore, 1 3 10 9 Y. enterocolitica
YadAwt were incubated in 1 ml of 50-fold diluted factor H or SC5b-9 Ab
at 4˚C for 1 h. Bacteria were spun down, and the supernatant was used for
detection of factor H at a final dilution of 1:300 or 1:100 for detection of
C5b-9 for flow cytometry, respectively. For Western blot analysis, the
factor H Ab was used at a final dilution of 1:1000. Peroxidase conjugated
F(ab)2 C3 Ab (goat–anti-human; 1:300) and FITC-conjugated F(ab)2 C3
Ab (goat–anti-human; 1:100) were obtained from Protos Immunoresearch.
The C3 Ab is described to recognize the C3 a-chain (∼120 kDa) and
b-chain (∼75 kDa), the C3b a9-chain (∼100 kDa), and the iC3b a9-chain
fragments (∼62 and ∼40 kDa). Purified IgG fraction was prepared from
polyclonal rabbit YadA antiserum (dilution 1:2000) that was collected
from immunized rabbits (7). Peroxidase-conjugated secondary anti-goat
Ab (1:5000) was obtained from Santa Cruz Biotechnology. DyLight488conjugated rabbit–anti-goat Ab (1:400) and APC-conjugated donkey–antirabbit Ab (1:200) were purchased from Jackson ImmunoResearch.
Materials and Methods
Binding of serum-derived C3 was investigated by incubation of Y. enterocolitica YadA0, Y. enterocolitica YadAwt, Y. enterocolitica G389A, and
Y. enterocolitica G389S (1 3 107) at 37˚C in 100 ml NHS (20%, 10%, 5%,
2.5%, 1.25%) or HIS diluted with PBS containing Mg2+ and Ca2+ (10%,
5%, 2.5%, 1.25%) for 30 min. Binding of purified proteins was analyzed
by incubation of bacteria in 100 ml of complement proteins C3, C3b, or
iC3b at several concentrations (20, 10, 5 mg/ml) in PBS. As a negative
control, each strain was treated with PBS only. After a washing step, the
samples were incubated with FITC-conjugated goat F(ab)2 anti-human C3
Ab. Following the Ab incubation, samples were washed once and finally
fixed in 1% PFA in PBS for flow cytometry analysis (BD Biosciences LSR
II). To detect the binding of serum-derived complement regulator factor
H, Y. enterocolitica YadA0, Y. enterocolitica YadAwt, Y. enterocolitica
G389A, and Y. enterocolitica G389S (1 3 107) were incubated at 37˚C in
100 ml HIS or HIDC3S (20%, 10%, 5%) diluted with PBS for 30 min. The
negative controls were samples incubated in PBS only. Samples were
washed and then incubated with polyclonal goat anti-human factor H Ab.
Bacterial strains and culture conditions
Y. enterocolitica YadA0 (25) was grown overnight in Luria–Bertani broth
at 27˚C supplemented with nalidixic acid (10 mg/ml) and kanamycin (50
mg/ml). Y. enterocolitica strains YadAwt, G389A, and G389S (5) were
grown in the presence of nalidixic acid, kanamycin, and spectinomycin (50
mg/ml). A 1:20 dilution of the overnight Y. enterocolitica culture was incubated for 1 h at 27˚C and an additional 2 h at 37˚C to induce expression
of YadA. The bacteria were washed once with PBS (Invitrogen), and the
OD at 600 nm was determined.
Sera, proteins, and Abs
Pooled normal human serum (NHS) from healthy donors was purchased
from the transfusion medicine department of the university hospital
Tübingen. C6-depleted serum (DC6S) and C3-depleted serum (DC3S) was
Sample preparation for Western blot analysis
Bacterial pellets were lysed in SDS sample buffer (1 3 108 Y. enterocolitica
cells were loaded per lane) and incubated for 5 min at 95˚C prior to
loading if not indicated otherwise.
Western blot analysis
After SDS-PAGE proteins were transferred onto nitrocellulose membranes,
the membranes were blocked for 1 h with PBS-5% milk powder at room
temperature. For the detection of C3 peroxidase conjugated goat F(ab)2
anti-human C3 Ab was used. Factor H was detected by a polyclonal rabbit
factor H antiserum and a peroxidase conjugated secondary anti-goat Ab.
Detection of bound Abs was performed using an enhanced chemiluminescence detection kit (Amersham Biosciences).
Complement deposition investigated by Western blot
To detect the binding of the complement factor C3 and C3 products
Y. enterocolitica YadA0, Y. enterocolitica YadAwt, Y. enterocolitica G389A,
and Y. enterocolitica G389S (1 3 109) were incubated at 37˚C in 200 ml of
50% HIS diluted with PBS. After 5 and 20 min, 100 ml of the samples
were removed and bacteria were separated from the supernatant by centrifugation (12,000 rpm for 1 min). Pellets were washed two times with
PBS 1% BSA and once with PBS. As an internal Ab specificity control,
one sample of each bacterial strain was incubated in PBS only. To analyze
the binding of purified C3b to Y. enterocolitica YadA0, Y. enterocolitica
YadAwt, Y. enterocolitica G389A, and Y. enterocolitica G389S (5 3 108)
bacteria were incubated at 37˚C in 100 ml of 40-mg/ml C3b diluted in PBS
for 30 min. Afterward, the bacteria were washed two times in PBS 1%
BSA and once in PBS. Finally, bacterial pellets were transferred into a new
tube and resuspended in sample buffer and suspected to SDS-PAGE and
Western blot.
Complement deposition analyzed by flow cytometry
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
cleavage product iC3b are recognized by phagocyte receptors
(CR1, CR3, CR4) (13). Finally, the binding of C3b to a C3 convertase changes this enzyme into a C5 convertase. This step is the
first toward the formation of the multiprotein complex termed
terminal complement complex (TCC; also known as the membrane attack complex). These complexes are pores with a diameter
of up to 10 nm that are stably inserted into the bacterial outer
membrane; they induce a loss of membrane integrity and lysis of
the pathogen (14).
To prevent complement-mediated self-destruction of host cells,
the entire complement system has to be tightly regulated at several
levels of the cascade. Key regulators of the complement cascade are
C4BP, factor H, factor H-like protein 1 (FHL1) and factor H-related
proteins (CFHR). C4BP accelerates the decay of the classical and
lectin C3 convertase and mediates degradation of C4b and also of
C3b (15, 16). Factor H, FHL-1, and CFHR5 accelerate the decay
of the inherently instable C3 convertases and act as cofactors for
factor I-mediated cleavage of C3b into iC3b and C3d (17, 18). A
number of pathogens express specific surface molecules to recruit
complement regulators like C4BP or factor H to the bacterial
surface. As a result, they can protect themselves from complement
attack (9, 10, 19).
The role of YadA for interaction with complement factors has
been investigated over the last few years. Skurnik and coworkers
compared the roles of Yersinia proteins YadA, Ail, and the LPS
O-antigen for complement resistance and concluded that YadA is
the most potent individual factor mediating resistance against
complement-mediated killing. In addition, they demonstrated that
surface-bound C3b levels do not correlate with a specific resistance phenotype, but that most strains resistant to complementmediated lysis accumulated iC3b on their surface (20–23). It was
shown that the stalk domain of the YadA trimer is involved in
preventing TCC killing (24). As a follow-up, Biedzka-Sarek et al.
(21) generated a large set of YadA stalk and neck deletion mutants to identify the exact binding domain for factor H. Their
data revealed that factor H targets several conformations of YadA
and discontinuous sites of the stalk. However, the consequences
of deletions in the coiled-coil stalk of YadA are not easily interpreted as distortions of the stalk may influence the structure of
distant parts of YadA.
In this study, we show that stable YadA trimer of Y. enterocolitica recruits C3bH2O and iC3b to the bacterial surface.
Binding of C3b then supports further recruitment of factor H,
which assists in the subsequent factor I-mediated conversion of
C3b to iC3b. In turn, this coating with iC3b prevents the formation
of TCC and thus protects Yersinia from complement-mediated
lysis. These data extend the current view of YadA-mediated complement evasion of Y. enterocolitica by adding a new mechanism
of factor H binding via the recruitment of C3b.
The Journal of Immunology
After a washing step, the samples were incubated with DyLight488conjugated rabbit–anti-goat Ab. After a washing step, all samples were
resuspended and fixed in 1% PFA in PBS and then analyzed in a flow
cytometer. To test how purified proteins C3b/iC3b might influence binding
of purified factor H, bacteria were preincubated with 300 mg/ml purified
C3b or iC3b for 30 min at room temperature. After a washing step, bacteria
were incubated with 100 mg/ml purified factor H for 30 min at room
temperature. Bacteria were washed again, and factor H that had bound to
the bacterial surface was subsequently stained as described above for detection by flow cytometry.
Serum killing assay
TCC formation investigated by flow cytometry
Formation of the TCC was investigated after incubation of Y. enterocolitica
YadA0, Y. enterocolitica YadAwt, (5 3 107) for 5, 10, and 15 min at 37˚C
in 100 ml of 25% or 50% NHS diluted with PBS. As a negative control,
each strain was treated with PBS only. After a washing step, the samples
were incubated with monoclonal rabbit–anti-human SC5b-9 Ab. The
samples were washed once and incubated with APC-conjugated goat–antirabbit Ab. After a washing step, all samples were resuspended and fixed in
1% PFA in PBS and then analyzed in a flow cytometer (BD Biosciences
LSR II).
deposition and whether YadA trimer stability may influence this
process. For this purpose, we used Y. enterocolitica strains expressing wild type YadA (YadAwt), YadA with reduced trimer
stability (YadA G389A and YadA G389S) (5, 7), or a strain deficient for YadA (YadA0). The strains were cultivated at 37˚C to
enforce YadA expression and surface levels of YadA were analyzed using specific Abs and flow cytometry (Fig. 1A). As expected, YadA0 strains did not express YadA while the YadA
surface levels of Y. enterocolitica G389A are comparable to those
of Y. enterocolitica YadAwt. Y. enterocolitica YadA G389S exhibits slightly reduced YadA surface levels (5). To study the role
of YadA in complement modulation, Y. enterocolitica strains were
incubated with NHS allowing complement activation on the surface, after which surface-bound C3 products was detected using
a polyclonal Ab against C3. In NHS, we found no significantly
different levels of surface-bound C3 products on all tested strains
(Fig. 1B). As a control, we incubated the strains with HIS in which
heat-labile complement factors like factor B and C2 are rendered
inactive (29). Because these factors are essential for convertase
formation, C3 activation and subsequent covalent C3b deposition
will not occur in the presence of HIS. Surprisingly, YadA expressing Y. enterocolitica strains showed high levels of surfacebound C3 products after incubation with HIS, whereas no C3
products were found on the surface of a YadA-negative strain (Fig.
1C, 1D). We observed that this binding was dose dependent and
that trimer stability was important for binding, because binding
was lower for Y. enterocolitica YadA G389A and YadA G389S.
Taken together, YadA expression fosters the binding of fluid-phase
C3 products from HIS to Y. enterocolitica, and trimeric stability of
YadA is associated with increased binding.
YadA binds to C3b and iC3b
Results
YadA mediates binding of C3 products to Y. enterocolitica
Y. enterocolitica strains expressing YadA are resistant to complement-dependent killing (26–28), and YadA was shown to bind
human complement inhibitors C4BP and factor H (20–23). However, the aim of this study was to find out how YadA controls C3b
FIGURE 1. Binding of C3 products to Y. enterocolitica. Y. enterocolitica strains were grown for
2 h at 37˚C to induce YadA-expression. (A) Histogram overlays of the YadA surface levels of all
bacterial strains used for the C3 binding assay.
(B–C) Aliquots of the same bacterial cultures were
used for incubations with different concentrations
of human serum. Bacteria were washed and C3
fragments bound to the bacterial surface were
stained with Abs specific for C3 (detecting all C3
products) and analyzed by flow cytometry. Yersinia
strains expressing wild type YadA (YadAwt), the
point mutated versions (YadA G389A or YadA
G389S) or a strain devoid of YadA (YadA0). (B)
Bacteria were incubated with NHS containing active complement. (C) Bacteria were incubated with
heat-inactivated serum (HIS) in which the complement activation cascade is not active. (D) Representative histogram overlays of (C) at 10% HIS.
Data are means of at least three individual experiments (B, C), or a representative experiment out of
three is shown (A, D).
To unravel which form of fluid-phase C3 (C3, C3b, iC3b) in HIS
is recruited by Y. enterocolitica YadA, we analyzed surface-bound
C3 by Western blotting (Fig. 2). To this end, Y. enterocolitica
YadAwt, Y. enterocolitica G389A, Y. enterocolitica G389S, and
Y. enterocolitica YadA0 were incubated in HIS and washed, and
the proteins of whole cell lysates were separated by SDS-PAGE.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
Y. enterocolitica were grown to exponential phase at 37˚C, washed once
with PBS and OD600 was determined; 3 3 106 bacteria were incubated in
25% DC6S for 45 min at 37˚C. After a washing step, the bacteria were
transferred into a new tube and were incubated in DC3S for 45 min at
37˚C. As controls, bacteria were incubated in heat-inactivated DC6S
(HIDC6S) followed by heat-inactivated DC3S (HIDC3S). Complement
activity was stopped by placing the samples on ice and by the addition of
1 volume of brain heart infusion medium. Serial dilutions (1022–1025) of
the bacteria were plated on selective agar and incubated at 27˚C for 24 h.
The serum bactericidal effect was calculated as the survival percentage,
taking the bacterial counts obtained with bacteria incubated in HIDC6S
and HIDC3S as 100%. The killing experiment was repeated for each strain
at least three times, starting from independent cultures.
3
4
YadA-MEDIATED COMPLEMENT EVASION
The separation of C3 products by SDS-PAGE allows discrimination
between C3 (a-chain ∼120 kDa and b-chain ∼5 kDa), C3b (a9-chain
∼100 kDa and b-chain ∼75 kDa) and iC3b (a9-chain fragments ∼62
kDa and ∼40 kDa chain, and b-chain ∼75 kDa; Fig. 2A). Owing to
covalent binding of the thioester to the bacterial surface, the 120 kDa
a9-chain of C3b and the 62 kDa a9-chain of iC3b will not be released
from the surface. The Western blot data indicate that YadA mediates
recruitment of C3b and iC3b from HIS (Fig. 2B). Accordingly,
Y. enterocolitica YadA0 did not exhibit any binding, and Y. enterocolitica G389S (bearing significantly reduced YadA surface levels
compared with Y. enterocolitica YadAwt and Y. enterocolitica
G389A) displayed lower surface levels of iC3b compared with
Y. enterocolitica YadAwt and Y. enterocolitica G389A.
Subsequently, we analyzed the binding of purified proteins C3,
C3b, and iC3b to YadA on the bacterial surface. Binding of purified
C3 could not be detected with any strain (Fig. 2C, left panel).
However, purified C3b and especially iC3b were bound by all strains
except Y. enterocolitica YadA0 in a dose-dependent matter. The
overall amount of acquired C3b and iC3b was significantly lower in
Y. enterocolitica G389A and G389S compared with Y. enterocolitica
YadAwt. Furthermore, we studied binding of purified C3d and
showed no direct binding of C3d to YadA (data not shown). These
data suggest that YadA specifically binds C3b and iC3b, but not C3.
YadA does not cleave C3b into iC3b
The above data indicate that YadA directly binds C3b; therefore, we
were prompted to investigate whether YadA itself could catalyze
the cleavage of C3b into iC3b. This would be beneficial for the
bacterium because conversion of C3b into iC3b would stop both
the amplification loop and downstream formation of the TCC. To
test this hypothesis, bacteria grown as described were incubated
with purified C3b. Afterward, samples were subjected to Western
blot analysis with C3-specific antiserum recognizing both the aand b-chain of C3, C3b, and iC3b (Fig. 3). C3b was bound by
Y. enterocolitica strains, but the 40 kDa band of iC3b was not
detectable. This finding demonstrates that in the absence of additional serum factors, Y. enterocolitica is not able to convert C3b
into iC3b. Thus, YadA does not directly degrade C3b.
C3 products derived from HIS and purified iC3b support
binding of factor H to Yersinia in a YadA-dependent manner
Conversion of C3b into iC3b is normally catalyzed by factor I in the
presence of a cofactor like factor H. Because it was shown previously that YadA binds factor H, we wondered whether the YadAdependent recruitment of C3b and iC3b could influence the binding
of factor H. As C3b and iC3b are recruited from HIS in a YadAdependent manner, this might actually enhance the binding of
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 2. Binding of purified C3, C3b, and iC3b to Y. enterocolitica. (A) Schematic representation of C3 and its cleavage products C3b and iC3b (m.w.
of the uncleaved a- and b-chain and of the cleavage products a9 is indicated in brackets). (B) Y. enterocolitica strains were incubated in 50% HIS for 20 min
and subjected to SDS-PAGE and Western blotting. From left to right, the following samples were loaded: heat-inactivated serum (HIS), Y. enterocolitica
YadAwt (incubated in HIS, washed and then resuspended in Laemmli buffer to obtain whole cell lysates containing all surface-bound complement proteins),
Y. enterocolitica YadA G389A, Y. enterocolitica YadA G389S, and Y. enterocolitica YadA0, which were all treated as the wild type strain. To better visualize the band pattern obtained with different C3 products, we also separated and blotted purified C3, C3b, and iC3b protein (separate membrane with
annotations to identify the respective a- and b-chains and fragments on the right side of Fig. 2B). C3 bound to the bacterial surface was detected with an
antiserum reacting with the C3 a- and b-chain as well as the cleavage products a9 and the 62-kDa and 40-kDa fragments of iC3b. Boxes highlight the a9chain of C3b or the 40-kDa chain of iC3b. (C) Y. enterocolitica strains expressing YadAwt (filled squares), YadA G389A (filled triangels), YadA G389S
(open triangles), or deficient for YadA YadA0 (open diamond) were incubated with the indicated concentrations of purified C3, C3b, or iC3b. Afterward, the
bound proteins were analyzed with flow cytometry. Data are means of at least three individual experiments (C), or one representative experiment out of
three is shown (B).
The Journal of Immunology
FIGURE 3. Testing of C3b cleavage activity of Y. enterocolitica.
Y. enterocolitica strains were incubated with purified C3b, washed, and
subjected to SDS-PAGE and Western blot. Abs that detect all C3 products
were used. Recombinant C3, C3b, and iC3b were loaded as controls and
size markers. To detect low signal intensities, Western blots were additionally exposed for prolonged time (long exposure). One representative
experiment of three is shown.
FIGURE 4. Factor H binding to Y. enterocolitica
from HIS and DC3S. (A) Flow cytometry analysis of
factor H bound to the bacterial surface after incubation
with HIS (20%, 10%, 5%) or HIDC3S (20%, 10%, 5%),
washed, and stained with antiserum directed against
factor H. As a negative control, bacteria were incubated
in PBS only. (B) The Western blot shows several dilutions of NHS and DC3S stained with antiserum against
factor H which also detect the a- and b-isoforms of
the complement factor H-related protein CFHR-1. (C)
Bacteria were immunostained for the presence of factor
H after incubation with either HIS compared with
HIDC3S (left panel) or after incubation with purified
factor H compared with bacteria incubated with iC3b
and factor H (right panel). Data are means of at least
three individual experiments (A), or one representative
experiment out of three is shown (B, C). **p , 0.01 for
all indicated significances.
levels detected with the lowest serum concentration of HIS
(i.e., 1%). To rule out that this was due to reduced levels of factor
H in the HIDC3S, a Western blot (Fig. 4B) with the same Ab that
was used for flow cytometry was performed. The blot clearly
demonstrates that the total amount of factor H (and also the related
CFHR-1) present in HIS and HIDC3S is comparable. These data
suggest that factor H binding to YadA depends on the presence of
C3 products. To test this hypothesis, we analyzed binding of purified factor H to Y. enterocolitica YadAwt that were preincubated
with purified C3b or iC3b compared with bacteria that were
treated with buffer alone before incubation with purified factor H.
We found that preincubation with iC3b could enhance binding of
factor H (Fig. 4C). The addition of C3b did not increase the
binding of factor H, which is in concordance with our previous
findings suggesting that YadA binds more strongly to iC3b than
C3b.
Inactivation of C3b by Y. enterocolitica YadAwt results in
diminished activation of the terminal pathway
To decipher whether the YadA-mediated capture and inactivation
of C3b interfered with the terminal complement pathway, TCC
formation on the bacterial surface was quantified by flow cytometry
(Fig. 5). Bacteria were incubated with serum and stained with an
Ab recognizing a neoepitope of C9 only present in the polymeric
C5b-9 complex. These data show that the formation of the TCC
on the surface of Y. enterocolitica YadAwt is significantly lower
compared with Y. enterocolitica YadA0. This finding is consistent
with the higher susceptibility of Y. enterocolitica YadA0 to serum
killing. Finally, we developed an assay to determine whether YadA
recruitment of C3b/iC3b actually contributes to inhibition of serum
killing (Fig. 6). Bacteria were first incubated in human serum de-
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
factor H to Y. enterocolitica and may result in formation of a tripartite YadA-factor H-C3b complex. To this end, factor H binding
from HIS and HIDC3S to Y. enterocolitica was compared using flow
cytometry. All strains (Y. enterocolitica YadAwt, Y. enterocolitica
G389A, Y. enterocolitica G389S, and Y. enterocolitica YadA0)
bound factor H dose dependently from 5%, 2%, and 1% HIS (Fig.
4A). There was no significant difference in the overall amount of
surface-associated factor H at one discrete serum concentration in
Y. enterocolitica YadAwt compared with Y. enterocolitica G389A
and Y. enterocolitica G389S. Only Y. enterocolitica YadA0 showed
significantly less factor H binding.
Performing the same assay with HIDC3S, binding of factor H
was significantly reduced even at the highest serum concentration
(i.e., 5%) compared with full serum and was even lower than the
5
6
YadA-MEDIATED COMPLEMENT EVASION
FIGURE 5. Surface detection of C5b-9 after incubation of Y. enterocolitica in NHS by flow cytometry. Bacteria were incubated in 25% or 50%
NHS for 5, 10, and 15 min, washed, and stained with
anti–C5b-9. Histogram overlays are shown from one
representative experiment.
FIGURE 6. Two-step serum killing assay of Y. enterocolitica in DC6S
and DC3S. (A) Assay design. Y. enterocolitica strains were grown for 2 h at
37˚C to induce YadA expression. After incubation in DC6S, DC3S or
a sequential incubation in DC6S followed by DC3S bacteria were plated
and CFUs were counted. (B) Survival of Y. enterocolitica incubated in the
indicated types of serum. CFUs of samples incubated with heat-inactivated
sera were taken as 100% survival. Data are means of at least three individual experiments.
conclude, this experiment suggests that YadA-dependent recruitment of C3b/iC3b is important for serum resistance. In addition,
these data show that YadA trimer stability is important for efficient
complement inactivation.
Discussion
YadA, a major virulence determinant of Y. enterocolitica, was
previously shown to be the most important single factor mediating
complement resistance. Our study demonstrates a novel mechanism for complement resistance mediated by YadA (schematic
overview in Fig. 7), amending the known mechanism through
direct binding of factor H. Instead of directly binding to factor H,
we show that YadA recruits C3b or iC3b to attract large amounts
of factor H. Furthermore, we show that trimeric stability of YadA
is essential for complement resistance.
In our experiments using purified C3 products, we observed that
YadA specifically binds C3b and iC3b but not C3 or C3d. This
differential binding can be explained by the different conformations of these molecules (Fig. 8). When native C3 is cleaved
into C3b, its thioester-containing domain (C3d) becomes exposed
to react with the bacterial surface. This is accompanied by a large
conformational change in the C3 molecule, also exposing new
binding sites for complement receptors and other complement
proteins like factor B. The fact that we observe binding of purified
C3b and iC3b to YadA indicates that YadA noncovalently associates with these proteins. During the purification process of C3b/
iC3b, the thioester bond reacts with water and is therefore no
longer active. Because YadA does not bind C3, we believe that it
specifically binds the C3b/iC3b conformation. Owing to the difficulties to purify trimeric YadA in its native structure, it will be
challenging to map the decisive region for interaction with C3b/
iC3b. Even minor changes in the protein sequence will disrupt the
trimer or tertiary protein structure.
Analyzing the deposition of C3 products with our laboratory
strain Y. enterocolitica WA-314 serotype O:8, we found that C3
binding upon incubation in active NHS was independent from
YadA expression and stability. This is in contrast to previous
findings in which less C3b deposition was observed in a YadApositive strain compared with YadA-negative bacteria (26). Our
finding might be contributable to the usage of diverse serotypes
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
ficient for C6 (DC6S). Here, binding of C3 products may occur in
the same way as in NHS, but the TCC cannot be assembled because of the lack of C6. After a washing step, bacteria were
transferred to human serum deficient for C3 (DC3S). As a result,
only C3 molecules that were bound in the first incubation step can
contribute to TCC formation. TCC-mediated killing of the bacteria
was monitored by plating serial dilutions (Fig. 6A). Neither of the
individually depleted sera induced bacterial lysis and thus both sera
were suitable for our purpose. However, sequential incubation of
bacteria in DC6S followed by DC3S serum triggered bacterial lysis
(Fig. 6B). Y. enterocolitica YadAwt, the strain that recruited the
greatest amount of C3b/iC3b (Fig. 2B), was most resistant in this
assay. Y. enterocolitica YadA G389A, G389S, and Y. enterocolitica
YadA0 exhibited increased susceptibility to serum killing. To
The Journal of Immunology
7
(26, 27, 30), individual serum activity, and strain-specific expression levels of factors involved in serum resistance (YadA, Ail
LPS), and also to subtle serotype-specific differences in the YadA
protein sequence. Recently, Ho et al. (31) showed that the outer
membrane protein Ail of Yersinia pseudotuberculosis binds to the
complement regulator C4BP regardless of serotype. However,
serum resistance was also independent of YadA in two strains of
FIGURE 8. Conformational changes of C3. Schematic drawing based on
protein structural information showing the gross conformational rearrangements upon cleavage of C3 in C3b/C3a; conversion of C3b in iC3b
and in C3dg/C3c. The cleavage of C3b into iC3b renders the molecule
into a more open conformation. Adapted by permission from Macmillan
Publishers Ltd: Gros, P., F. J. Milder, and B. J. Janssen. 2008. Complement
driven by conformational changes. Nat. Rev. Immunol. 8: 48–58, copyright
2008.
Y. pseudotuberculosis, whereas in another strain serum resistance
involved factors in addition to YadA and Ail. These data demonstrate that a complex interplay between bacterial proteins involved in serum resistance defines the resistance phenotype and
that YadA in contrast to Ail in some cases seems dispensable for
serum resistance.
Our finding that YadA binds to C3b/iC3b was initially triggered
by experiments in which we incubated YadA-expressing bacteria
with HIS. The purpose of heat-treatment of serum at 56˚C is to
inactivate the most heat-labile complement components being
C2, C1, and factor B (29, 32). As a consequence, the complement
cascade cannot be initiated and C3 will not be cleaved into C3b.
Although we expected that HIS only has C3 and none of its activation products, we observed that our HIS sample contained high
amounts of C3b and iC3b (Fig. 2B). These C3 products were
probably generated during heat inactivation. Of course, bacterial
pathogens will not encounter HIS in vivo, but the finding is still
physiologically relevant because C3 products such as C3b and
iC3b are continuously generated during complement activation
on the bacterial surface. The advantage of using heat-inactivated
serum is that it allowed us to examine the recruitment of C3 products in the absence of covalent C3b deposition.
Our study challenges the current model that YadA directly binds
to factor H. With YadA being expressed on the bacterial surface, we
did not observe direct binding of factor H to YadA. Factor H is
known to bind to C3b and iC3b, but not to C3 (33), and it therefore
seems likely that it is bound to YadA via C3b/iC3b. Unfortunately,
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 7. Model of YadA-mediated complement evasion. (A) Complement is activated by the classical pathway (CP), the lectin pathway (LP), or the
alternative pathway (AP), which leads to the opsonization with C3b and/or C4b (not shown). (B) Current direct binding model of factor H-mediated
complement evasion. Factor H binds directly to YadA (BI) and binds to C3b deposited on the bacterial surface (BII). This binding enables factor I to cleave
C3b into iC3b (BIII) and thereby restricts the late steps of the complement cascade. (C) New C3b recruitment model of factor H binding amending the current
direct binding model. In a wild type strain of Y. enterocolitica, YadA mediates recruitment of C3b (CI). Subsequently, C3b enables efficient binding of factor
H (CII), which together with factor I leads to inactivation of C3b into iC3b (CIII). iC3b then fosters binding of factor H to C3b (CIV) and the amplification of
factor H/factor I-mediated generation of iC3b (CV). Thus, the formation of TCC is limited (CVI). The binding sites for C3b and factor H within YadA are
unknown (C3b) or cannot be assigned to one distinct region (factor H). The models do not reflect information about C3b and factor H binding sites.
8
Hia, or UspA1 (39), or small deletions within the YadA stalk domain
(21) significantly reduce the serum resistance of Y. enterocolitica.
In the current study, we demonstrated that Y. enterocolitica
recruits C3b/iC3b via YadA to attract large amounts of factor H
and by this mechanism significantly contributes to Y. enterocolitica complement resistance. Our work emphasizes the importance of YadA as a pathogenicity factor that does not only
mediate adherence but also efficient immune evasion. As a result,
YadA is an attractive target for the development of anti-infectives
or the creation of vaccines. Given that YadA represents a large
family of proteins that share common structural characteristics, it
remains to be elucidated whether our findings are assignable to
other pathogenic species. Future studies might succeed in mapping the exact interaction sites of YadA and C3b/iC3b to enable
interference with this interaction. In vivo infection models will
then reveal the physiologic relevance of YadA-mediated recruitment of C3b/iC3b.
Disclosures
The authors have no financial conflicts of interest.
References
1. Bottone, E. J. 1999. Yersinia enterocolitica: overview and epidemiologic correlates. Microbes Infect. 1: 323–333.
2. Cover, T. L., and R. C. Aber. 1989. Yersinia enterocolitica. N. Engl. J. Med. 321:
16–24.
3. Cornelis, G. R., A. Boland, A. P. Boyd, C. Geuijen, M. Iriarte, C. Neyt,
M. P. Sory, and I. Stainier. 1998. The virulence plasmid of Yersinia, an antihost
genome. Microbiol. Mol. Biol. Rev. 62: 1315–1352.
4. El Tahir, Y., and M. Skurnik. 2001. YadA, the multifaceted Yersinia adhesin. Int.
J. Med. Microbiol. 291: 209–218.
5. Schütz, M., E. M. Weiss, M. Schindler, T. Hallström, P. F. Zipfel, D. Linke, and
I. B. Autenrieth. 2010. Trimer stability of YadA is critical for virulence of
Yersinia enterocolitica. Infect. Immun. 78: 2677–2690.
6. Linke, D., T. Riess, I. B. Autenrieth, A. Lupas, and V. A. Kempf. 2006. Trimeric
autotransporter adhesins: variable structure, common function. Trends Microbiol.
14: 264–270.
7. Grosskinsky, U., M. Schütz, M. Fritz, Y. Schmid, M. C. Lamparter, P. Szczesny,
A. N. Lupas, I. B. Autenrieth, and D. Linke. 2007. A conserved glycine residue
of trimeric autotransporter domains plays a key role in Yersinia adhesin A
autotransport. J. Bacteriol. 189: 9011–9019.
8. Lambris, J. D., D. Ricklin, and B. V. Geisbrecht. 2008. Complement evasion by
human pathogens. Nat. Rev. Microbiol. 6: 132–142.
9. Rooijakkers, S. H., and J. A. van Strijp. 2007. Bacterial complement evasion.
Mol. Immunol. 44: 23–32.
10. Blom, A. M., T. Hallström, and K. Riesbeck. 2009. Complement evasion strategies of pathogens-acquisition of inhibitors and beyond. Mol. Immunol. 46:
2808–2817.
11. Zipfel, P. F., R. Würzner, and C. Skerka. 2007. Complement evasion of pathogens: common strategies are shared by diverse organisms. Mol. Immunol. 44:
3850–3857.
12. Gros, P., F. J. Milder, and B. J. Janssen. 2008. Complement driven by conformational changes. Nat. Rev. Immunol. 8: 48–58.
13. Kemper, C., and J. P. Atkinson. 2007. T-cell regulation: with complements from
innate immunity. Nat. Rev. Immunol. 7: 9–18.
14. Tegla, C. A., C. Cudrici, S. Patel, R. Trippe, III, V. Rus, F. Niculescu, and
H. Rus. 2011. Membrane attack by complement: the assembly and biology of
terminal complement complexes. Immunol. Res. 51: 45–60.
15. Blom, A. M., L. Kask, and B. Dahlbäck. 2003. CCP1-4 of the C4b-binding
protein alpha-chain are required for factor I mediated cleavage of complement
factor C3b. Mol. Immunol. 39: 547–556.
16. Blom, A. M., B. O. Villoutreix, and B. Dahlbäck. 2004. Functions of human
complement inhibitor C4b-binding protein in relation to its structure. Arch.
Immunol. Ther. Exp. (Warsz.) 52: 83–95.
17. Józsi, M., and P. F. Zipfel. 2008. Factor H family proteins and human diseases.
Trends Immunol. 29: 380–387.
18. Zipfel, P. F., and C. Skerka. 2009. Complement regulators and inhibitory proteins. Nat. Rev. Immunol. 9: 729–740.
19. Zipfel, P. F. 2009. Complement and immune defense: from innate immunity to
human diseases. Immunol. Lett. 126: 1–7.
20. Biedzka-Sarek, M., H. Jarva, H. Hyytiäinen, S. Meri, and M. Skurnik. 2008.
Characterization of complement factor H binding to Yersinia enterocolitica serotype O:3. Infect. Immun. 76: 4100–4109.
21. Biedzka-Sarek, M., S. Salmenlinna, M. Gruber, A. N. Lupas, S. Meri, and
M. Skurnik. 2008. Functional mapping of YadA- and Ail-mediated binding of
human factor H to Yersinia enterocolitica serotype O:3. Infect. Immun. 76: 5016–
5027.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
it was not possible to restore the binding of factor H by reconstituting the C3-depleted serum with purified C3, C3b, or iC3b.
The conformation of the proteins is possibly altered during the
purification process in a way that abrogates efficient interaction
with factor H. In addition, our finding raises the question about the
biologic relevance of enhanced factor H binding. It has been
shown that factor H, which binds directly to YadA, acts as a cofactor for factor I-mediated degradation of C3b. However, these
data also revealed that only a very small amount of C3b present in
the experiments was degraded into iC3b (5, 20, 21). The amount
of C3b typically used in these experiments was considerably lower
than that of C3 in human serum. Thus, it can be speculated that,
especially in vivo, the amount of factor H that binds directly to
YadA is not sufficient to degrade all bound C3b-molecules and
Y. enterocolitica rely on additional factor H binding mediated by
C3b/iC3b.
Active binding and inactivation of C3b is a two-edged sword. On
the one hand, the rapid inactivation of C3b on the bacterial surface
is advantageous for Y. enterocolitica. The binding and inactivation
of C3b by factor H and factor I might lower the local amount of
C3 in the fluid phase and may form a kind of “protective shell”
that sterically hinders further deposition of C3b. In addition, iC3b
stabilizes the complex of C3b and factor H (34), which in turn
enhances inactivation of C3b into iC3b. Taken together, these
effects might inhibit the complement cascade at the stage of C3
and prevent later steps of the complement cascade. Indeed, a reduced number of TCC on Y. enterocolitica YadAwt compared with
YadA-deficient bacteria could be observed as it was described
earlier (27). On the other hand, iC3b is a very potent opsonin, has
high affinity and avidity to complement receptors expressed on
phagocytes, induces a strong oxidative burst, and is thus detrimental for the pathogen (35). However, Y. enterocolitica encodes
for a type 3 secretion system that injects host cells with several
effector proteins that induce the rearrangement of the host cell
cytoskeleton and thus prevent the internalization of Y. enterocolitica (36). One could envision a scenario in which Y. enterocolitica exploits opsonins to tightly adhere to target cells. Because
adherence to host cells is the prerequisite for efficient effector
injection, it might be enhanced by surface deposited C3b/iC3b and
its interaction with complement receptors CR1, 3, and 4 (37).
However, the interaction of Y. enterocolitica with its target cells is
a highly complex and multifactorial process, and further investigation has to reveal the effects of complement interaction on the
pathogenesis of Y. enterocolitica.
The mechanism presented in this article for YadA shows strong
similarities with the Staphylococcus aureus protein Sbi that forms
a tripartite complex with C3b and the complement regulator factor
H. Moraxella catharralis, which expresses the trimeric autotransporter adhesins UspA1 and 2 (that share structural homologies to
YadA), can noncovalently attach to C3 and methylamine-treated
C3 (this does not offer a thioester bond for attachment, but otherwise resembles C3b) (38).
Our studies strongly indicated that the trimeric stability of YadA
is important for its interaction with C3b/iC3b. The significant
differences observed in C3b/iC3b binding between Y. enterocolitica
YadAwt and Y. enterocolitica G389A and Y. enterocolitica G389S
might be explained by the fact that even point mutations within
the membrane-embedded C-terminal b-barrel-domain can lead to
distortions at very distant parts of the coiled-coil stalk region.
These distortions might then adversely affect the interplay between YadA and complement regulatory factors. Indeed, it has
been reported earlier that the exchange of the entire YadA membrane anchor domain against the C-terminal domains of other
trimeric autotransporter adhesin family members such as EibA,
YadA-MEDIATED COMPLEMENT EVASION
The Journal of Immunology
31. Ho, D. K., R. Riva, V. Kirjavainen, H. Jarva, E. Ginström, A. M. Blom,
M. Skurnik, and S. Meri. 2012. Functional recruitment of the human complement inhibitor C4BP to Yersinia pseudotuberculosis outer membrane protein
Ail. J. Immunol. 188: 4450–4459.
32. Mayer M. M., ed. 1961. Complement and complement fixation. Charles
C Thomas, Springfield.
33. Wu, J., Y. Q. Wu, D. Ricklin, B. J. Janssen, J. D. Lambris, and P. Gros. 2009.
Structure of complement fragment C3b-factor H and implications for host protection by complement regulators. Nat. Immunol. 10: 728–733.
34. Kajander, T., M. J. Lehtinen, S. Hyvärinen, A. Bhattacharjee, E. Leung,
D. E. Isenman, S. Meri, A. Goldman, and T. S. Jokiranta. 2011. Dual interaction
of factor H with C3d and glycosaminoglycans in host-nonhost discrimination by
complement. Proc. Natl. Acad. Sci. USA 108: 2897–2902.
35. Ehlers, M. R. 2000. CR3: a general purpose adhesion-recognition receptor essential for innate immunity. Microbes Infect. 2: 289–294.
36. Cornelis, G. R. 2002. Yersinia type III secretion: send in the effectors. J. Cell
Biol. 158: 401–408.
37. Underhill, D. M., and A. Ozinsky. 2002. Phagocytosis of microbes: complexity
in action. Annu. Rev. Immunol. 20: 825–852.
38. Nordström, T., A. M. Blom, T. T. Tan, A. Forsgren, and K. Riesbeck. 2005.
Ionic binding of C3 to the human pathogen Moraxella catarrhalis is a
unique mechanism for combating innate immunity. J. Immunol. 175: 3628–
3636.
39. Ackermann, N., M. Tiller, G. Anding, A. Roggenkamp, and J. Heesemann. 2008.
Contribution of trimeric autotransporter C-terminal domains of oligomeric
coiled-coil adhesin (Oca) family members YadA, UspA1, EibA, and Hia to
translocation of the YadA passenger domain and virulence of Yersinia enterocolitica. J. Bacteriol. 190: 5031–5043.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
22. Biedzka-Sarek, M., R. Venho, and M. Skurnik. 2005. Role of YadA, Ail, and
Lipopolysaccharide in Serum Resistance of Yersinia enterocolitica Serotype O:3.
Infect. Immun. 73: 2232–2244.
23. Kirjavainen, V., H. Jarva, M. Biedzka-Sarek, A. M. Blom, M. Skurnik, and
S. Meri. 2008. Yersinia enterocolitica serum resistance proteins YadA and ail
bind the complement regulator C4b-binding protein. PLoS Pathog. 4: e1000140.
24. Roggenkamp, A., N. Ackermann, C. A. Jacobi, K. Truelzsch, H. Hoffmann, and
J. Heesemann. 2003. Molecular analysis of transport and oligomerization of the
Yersinia enterocolitica adhesin YadA. J. Bacteriol. 185: 3735–3744.
25. Roggenkamp, A., H. R. Neuberger, A. Flügel, T. Schmoll, and J. Heesemann.
1995. Substitution of two histidine residues in YadA protein of Yersinia enterocolitica abrogates collagen binding, cell adherence and mouse virulence. Mol.
Microbiol. 16: 1207–1219.
26. China, B., M. P. Sory, B. T. N’Guyen, M. De Bruyere, and G. R. Cornelis. 1993.
Role of the YadA protein in prevention of opsonization of Yersinia enterocolitica
by C3b molecules. Infect. Immun. 61: 3129–3136.
27. Pilz, D., T. Vocke, J. Heesemann, and V. Brade. 1992. Mechanism of YadAmediated serum resistance of Yersinia enterocolitica serotype O3. Infect. Immun.
60: 189–195.
28. Pilz, D., E. Wachter, J. Heesemann, and V. Brade. 1991. Evasion of Yersinia
enterocolitica serotype 03 from complement-mediated killing. Behring Inst.
Mitt. 183–193.
29. Joisel, F., I. Leroux-Nicollet, J. P. Lebreton, and M. Fontaine. 1983. A hemolytic
assay for clinical investigation of human C2. J. Immunol. Methods 59: 229–235.
30. Tertti, R., E. Eerola, O. P. Lehtonen, T. H. Ståhlberg, M. Viander, and
A. Toivanen. 1987. Virulence-plasmid is associated with the inhibition of
opsonization in Yersinia enterocolitica and Yersinia pseudotuberculosis. Clin.
Exp. Immunol. 68: 266–274.
9