Vaccine 27 (2009) 3276–3280
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/vaccine
Heterogeneous in vivo expression of clumping factor A and capsular
polysaccharide by Staphylococcus aureus: Implications for vaccine design
Jasdeep S. Nanra, Yekaterina Timofeyeva, Sandra M. Buitrago, Bret R. Sellman1, Deborah A. Dilts,
Pamela Fink, Lorna Nunez, Michael Hagen, Yury V. Matsuka, Terri Mininni, Duzhang Zhu,
Viliam Pavliak, Bruce A. Green, Kathrin U. Jansen, Annaliesa S. Anderson∗
Wyeth Vaccine Research, 401 N. Middletown Road, Pearl River, NY 10965, USA
a r t i c l e i n f o
Available online 5 February 2009
a b s t r a c t
There is a clear unmet medical need for a vaccine that would prevent infections from Staphylococcus
aureus (S. aureus). To validate antigens as potential vaccine targets it has to be demonstrated that the
antigens are expressed in vivo. Using murine bacteremia and wound infection models, we demonstrate
that the expression of clumping factor A (ClfA) and capsular polysaccharide antigens are heterogeneous
The data presented in this report support a multiantigen approach for the development of a prophylactic
S. aureus vaccine to ensure broad coverage against this versatile pathogen.
© 2009 Elsevier Ltd. All rights reserved.
normally colonize humans without causing disease. When disease
occurs it can range from mild skin infections to more devastat-
ing diseases such as bacteremia, osteomyelitis and endocarditis.
in both hospital and community settings have reduced treatment
options and necessitated a need for alternative approaches. One
such approach is to develop a prophylactic vaccine. Since humans
are exposed to S. aureus, the immune system comes into frequent
contact with the organism and associated potential antigens. This
is exemplified by immunological surveys that have looked for and
found anti-staphylococcal antibodies in humans . There are sev-
eral lines of indirect evidence supporting the role of antibody
mediated opsonophagocytosis as being important for host defense
susceptible to S. aureus infections . The complement system in
conjunction with antibodies plays an important role in opsoniz-
ing bacteria as demonstrated in both in vitro and in vivo settings
[4,5] and function to target the bacteria for destruction by poly-
∗Corresponding author. Tel.: +1 845 602 4674; fax: +1 845 602 4941.
E-mail address: email@example.com (A.S. Anderson).
1Present address: Medimmune, Inc., Department of Infectious Diseases, 1 Med-
immune Way, Gaithersburg, MD 20878, USA.
morphonuclear neutrophil cells (PMNs). In principle, a S. aureus
vaccine would work by eliciting functional antibodies that bind
the target antigen(s) on the bacterial surface and trigger com-
plement activation, which in turn would augment the process of
uptake and killing of bacteria by phagocytic cells such as neu-
Currently, no licensed prophylactic vaccine exists although clin-
ical trials have and are being conducted to identify such a vaccine
. A successful S. aureus vaccine should be composed of con-
served antigen(s) that (a) are expressed on the cell surface of
the bacteria in vivo and therefore accessible to antibodies, (b)
are detectable on relevant clinical isolates during infection and
(c) elicit functional antibody responses. For S. aureus the pro-
cess of choosing the vaccine antigen(s) is problematic due to the
complex biology of this organism, the broad spectrum of dif-
ferent diseases that it causes, the complex regulatory pathways
that govern the expression of surface antigens that can differ
from strain to strain and its ability to evade the immune sys-
Among the most widely investigated S. aureus antigens are
capsular polysaccharides. They are virulence factors that inhibit
opsonophagocytic killing by neutrophils . The predominant cap-
sular polysaccharides types are 5 and 8, which are expressed by
∼85–90% of S. aureus clinical isolates . Extensive preclinical data
showing the efficacy of a CP5 conjugate vaccine in different ani-
mal models have been published [9–11] and human antibodies
generated using a CP5 conjugate vaccine are capable of protect-
ing mice from infection and death . It is of note that there are
0264-410X/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.
J.S. Nanra et al. / Vaccine 27 (2009) 3276–3280
few published reports demonstrating preclinical efficacy for a CP8
conjugate vaccine .
In addition to capsular polysaccharides, several S. aureus pro-
tein antigens have also been tested preclinically with promising
results. Microbial surface components recognizing adhesive matrix
molecules (MSCRAMMs) are adhesins involved in staphylococci
cal implants . It has been demonstrated that adhesins including
clumping factor A (ClfA), fibronectin binding protein A (FnbpA) and
iron surface determinant B (IsdB) are efficacious vaccine antigens
in different animal models [14–16] and IsdB is currently in phase II
Despite the encouraging preclinical data for a CP conjugate vac-
cine, no efficacy was observed in a clinical trial in an end stage
renal dialysis (ESRD) patient population . A potential reason
for the failure may be the heterogeneity of capsule expression
within the infecting S. aureus clinical strains. As stated earlier,
for an antigen to be an effective vaccine component it must be
expressed in vivo. There are limited published studies examin-
ing CP expression in vivo and in those studies expression was not
monitored during the initial stage of infection (<6h post infection)
[18–20]. This early stage is when antibody mediated immunity has
the greatest chance of being effective prior to the formation of
The expression of target antigens and their accessibility to anti-
bodies in vivo are important considerations for predicting whether
a target will succeed and for an understanding of possible failure.
Directing efforts towards addressing these questions is important
to develop successful vaccines to prevent S. aureus disease. In this
paper, we examine the in vivo expression kinetics of CP and ClfA
in murine bacteremia and wound infection models. The expression
of both antigens was dependent on the challenge strain and the in
2. Materials and methods
2.1. Immunofluorescent microscopy
2.1.1. S. aureus isolation from infected animals.
Bacteremia infection model. Female 12-week-old CD1 mice
(Charles River, MA) were infected by intraperitoneal injection
(500?l) of S. aureus grown to OD600nm∼2.5 in tryptic soy broth
(TSB) (Becton, Dickinson and Co., CA). Infected mice were exsan-
guinated and the blood from three mice was pooled into ice-cold
sodium citrate buffer (final concentration of sodium citrate 0.4%,
pH 7.0). The blood cells were lysed with 1% NP-40 (Pierce Biotech-
nology) and recovered bacteria and cells were washed twice with
phosphate buffered saline (PBS, pH 7.4, Mediatech., Inc.). A 100?l
aliquot of a bacterial suspension in PBS was added to Lab-Tek®II
CC2 Chamber Slides (Thermo Fisher Scientific, Rochester, NY) for
Wound infection model. 6–8-week-old C57BL/6 female mice
(Charles River, MA) underwent surgery to embed a loop of suture
into a thigh muscle incision. Five microliter of S. aureus grown to
OD600nm∼2.5 in TSB (∼5×106cfu) were introduced into the inci-
sion under the deep tissue suture. The skin was then closed with
skin closure clips. Infected animals were euthanized, the wounds
were opened, and bacterial samples were obtained by swabbing.
Swab samples were added to Lab-Tek®II CC2 Chamber Slides.
Staining was performed on Lab-Tek®II CC2 Chamber Slides.
Rabbit anti-CP5, anti-CP8, and anti-ClfA affinity purified IgGs
(1microgram/ml conc.) were used as primary antibodies for
immunostaining. Bacteria were incubated overnight at 4◦C with
rabbit primary antibodies, cells were then washed three times
in PBS and stained with Alexa488 conjugated goat-anti rab-
bit antibody (1:250) (Invitrogen, Carlsbad, CA). The labeled
bacteria were coated with BioMedaTM
Corporation, Foster City, CA). Images were obtained with a
Leica TCS SL spectral confocal microscope (Leica Microsys-
2.3. Flow cytometry
Monoclonal antibodies (mAbs) CP5-7-1, CP8-3-1 and 12-9 spe-
cific for CP5, CP8 and ClfA respectively, were used as primary
antibodies for flow cytometry. 12-9 was kindly provided by Inhib-
itex (Alpharetta, GA) . CP5-7-1 and CP8-3-1 were isolated
in-house using standard protocols . In two-color flow analy-
sis, mAbs 12-9 and CP5-7-1 were conjugated to fluorophores using
an Alexa488 and an Alexa633 protein binding kit respectively, as
per manufacturers instructions (Invitrogen, Carlsbad, CA). CP8-3-
1-Alexa633 conjugate and isotype-matched mouse IgG1-Alexa488
conjugate were included as staining controls. To initiate a bac-
terial culture, a frozen stock of strain Reynold was subcultured
onto tryptic soy agar (Becton Dickinson and Co., CA) and incubated
at 37◦C overnight. A single colony was inoculated into 20ml of
Columbia broth (Becton Dickinson and Co., CA) with 2% sodium
chloride, shaken at 150rpm, and incubated at 37◦C overnight.
Bacterial cells were centrifuged and cell pellet resuspended in
Hank’s balanced salt solution (Mediatech Inc., VA) and this step
was repeated twice. Protein A on the bacterial surface was blocked
by a 30min preincubation on ice with staining buffer contain-
ing 10% porcine serum. Bacteria (∼107cfu) were then incubated
with primary antibodies (10?g/ml conc.) in staining buffer for
30min on ice. Cells were then washed twice in staining buffer
and stained with FITC-conjugated goat anti-mouse IgG (Jackson
staining buffer, fixed with 2% paraformaldehyde (v/v) in PBS. Data
were acquired using a BD LSRII flow cytometer (Becton Dickinson
and Co., CA) and analyzed using the FlowJo software (Tree Star
Inc., OR). A total of 20,000 events were collected for each sam-
2.4. Opsonophagocytic assay
HL-60 cells (ATCC, VA) were differentiated using dimethylfor-
mamide (Fisher Scientific, NJ) as described by Romero-Steiner et al
. Harvested HL-60 cells were resuspended in Hanks balanced
salt solution with 0.1% gelatin (HBG assay buffer) at ∼108cells/ml
concentration and placed in a 37◦C incubator until ready to use.
S. aureus strain Reynold was grown overnight in Columbia broth
with 2% sodium chloride. Bacterial cells were washed twice with
phosphate buffered saline pH 7.0 (Mediatech, VA), resuspended in
assay buffer to a concentration of 106cfu/ml and placed on ice. S.
aureus CP5 or CP8 specific monoclonal antibodies, affinity purified
human antibodies to ClfA or an irrelevant antigen (streptococcal
C5a peptidase (SCP) from Streptococcus pyogenes, negative control)
was tested in the assay. Affinity purified human IgGs and the mon-
oclonal antibodies were adjusted to a concentration of 65?g/ml
and 10?l were added to a 96 well plate (Corning Inc., NY) contain-
ing the assay buffer (60?l). Bacteria were added to the mixture
(10?l) and incubated at 4◦C on platform shaker (250rpm). Serum
from human donors adsorbed against S. aureus type 5, type 8 and
non-typeable strains was used as a source of complement in the
assay. Differentiated HL-60 cells (10?l) were added at an effec-
tor:target cell ratio of 100:1, and the suspension was mixed well by
repeated pipetting. The assay plate was incubated at 37◦C for 1h
J.S. Nanra et al. / Vaccine 27 (2009) 3276–3280
In vivo expression of clumping factor A and capsular polysaccharide in type 5 S. aureus isolates.
by majority of the bacterial cells. d: “±” denotes that at least 50% of the bacterial cells were expressing the
antigen. e: “−” denotes undetectable antigen expression on bacterial cells.
on a platform shaker (250rpm). A 10?l aliquot of the reaction mix
was serially diluted 10 fold in a sterile 1% (w/v) saponin solution
in water (Sigma, MO) and 10?l dribbled on tryptic soy agar plates
in duplicate. The percentage killing was calculated by determining
the ratio of the number of colony forming units (CFU) surviving at
60min in wells with bacteria, antibodies, complement and HL-60
cells to the number of CFU surviving in tubes lacking antibod-
ies but containing bacteria, complement and HL-60 cells. Controls
containing bacteria, complement and monoclonal antibody were
included to adjust for any reduction in CFU due to bacterial clump-
3. Results and discussion
Using an immunofluorescent microscopy technique, we moni-
tored the in vivo expression of CP5/8 and ClfA in murine bacteremia
and wound infection models. Several CP type 5 and 8 clinical iso-
lates were used as challenge strains. Upon entry into the host, for
most of the CP type 5 strains, the expression of both CP5 and ClfA
was below the limit of detection at the 1h time point. At later time
points the expression of these antigens became visible but varied
and was dependent on the strain and the infection model (Table 1).
All of the tested CP5 strains expressed ClfA in both models, but the
Fig. 1. Surface expression of capsular polysaccharide type 5 (CP5) and clumping factor A on S. aureus Reynold using flow cytometry. The bacteria were grown overnight in
Columbia broth with 2% NaCl. (A) Bacteria were stained with CP5-7-1 and 12–9mAbs (filled graphs) and compared to staining with isotype matched CP8-3-1 specific mAb
(open graphs). (B) A two-color flow analysis staining was done using Alexa488 conjugated 12-9 and Alexa633 conjugated CP5-7-1mAbs and compared to staining with
Alexa488 conjugated isotype matched control mouse IgG and Alexa633 conjugated CP8-3-1mAb. A total of 20,000 events were collected for each sample. These data are
representative of two independent experiments.
J.S. Nanra et al. / Vaccine 27 (2009) 3276–3280
In vivo expression of clumping factor A and capsular polysaccharide in type 8 S. aureus isolates.
timing of expression varied by strain. Expression of CP5 was also
observed in both the bacteremia and wound models but with dif-
ferent expression kinetics. All of the tested type 5 strains expressed
capsule at some point in the bacteremia model, and 5 of 6 strains
expressed CP5 in the wound model. One isolate from the USA300
lineage (CDC 3) was also tested and expressed CP5 only under in
vivo conditions in both models (Table 1). CP5 expression was not
detectable on the CDC3 isolate grown under in vitro conditions. A
Other staphylococcal antigens such as IsdB  and SdrG, a S. epi-
dermidis protein , have also been shown to be dependent on
an in vivo environment for their expression. In the study by Sell-
man et al. , SdrG expression was undetectable on S. epidermidis
grown in tryptic soy broth or 70% human sera. However, bacteria
the expression of this antigen in the tested strains.
In vivo expression of CP and ClfA was also quite variable for the
type 8 strains. All the type 8 strains tested expressed both cap-
sule and ClfA at the time of challenge. Under in vivo conditions
however, these antigens were heterogeneously expressed and the
expression was again dependent on the strain, time point, and the
ClfA and CP early in the blood compared to the wound environ-
ment and this expression pattern was unique to this strain. In 8
of 12 CP8 strains, the expression of capsular polysaccharide was
not observed at the 4h time point in either model. However, ClfA
expression was observed in 9 of 12 strains at the 4h time point.
Our data highlight the strain dependent differences in CP and ClfA
in vivo expression profiles. These results also demonstrate that in
S. aureus CP and ClfA expression is differentially regulated depend-
ing on the in vivo microenvironment, i.e., site of infection. These
results may also explain the variable efficacy results reported for
CP8 conjugates in animal models. In a recent report, a CP8 conju-
gate vaccine had shown inconsistent efficacy in the murine lethal
challenge model . Our in vivo expression data (Table 2) demon-
strate that most of the tested type 8 isolates do not express CP in
the blood till later time points post-challenge (>4h), which could
explain the reported inconsistent results.
It is also evident from the in vivo expression studies that ClfA is
accessible to antibodies in the presence of capsule expression. To
substantiate our in vivo findings that capsule expression does not
prevent antibodies from binding other vaccine target antigens, we
analyzed the binding of a ClfA-specific monoclonal antibody to S.
aureus Reynold grown under conditions (high salt concentration)
that enhance capsule expression (Fig. 1). ClfA and CP5 expression
could be detected on the surface of strain Reynold demonstrating
that antibodies to ClfA can bind to the bacterial target in the pres-
ence of high capsule expression (Fig. 1A). To demonstrate that both
ClfA and CP5 antigens were available for binding by antibodies on
the same bacterial cell a two-color flow cytometry was performed
(Fig. 1B). The results show that 95% of the bacteria expressed both
antigens and that the capsule expression did not prevent mAb 12-9
from binding ClfA. In addition to binding, the opsonic activity of
ClfA antibodies was also tested against the CP5 expressing strain
Reynold. Affinity purified human ClfA specific IgGs were opsonic
as measured by bacterial killing (Fig. 2) illustrating that these anti-
bodies can bind ClfA in the presence of capsular polysaccharide.
The presence of CP on the bacteria was also confirmed by opsonic
activity of CP5 antibodies. Specificity of the anti-ClfA and anti-CP5
opsonic activity was confirmed as the respective controls showed
sion on the bacterial surface.
cine will likely provide broad coverage against the majority of S.
aureus infections because of the heterogeneity of vaccine target
Fig. 2. Antibodies to ClfA induce opsonophagocytic killing of S. aureus expressing
capsular polysaccharide in opsonophagocytosis assays. Strain Reynold was grown
overnight in Columbia broth with 2% NaCl. Bacteria were opsonized with ClfA affin-
ity purified human IgG or irrelevant antigen affinity purified human IgG (negative
control, streptococcal SCP protein) and the opsonic activity tested. Differentiated
HL-60 cells were used in the opsonophagocytic assay at an effector/target cell ratio
of 100:1. As an additional control, a CP5 mAb was included in the experiment to
demonstrate the presence of CP5 on the surface. The results are average of two
independent experiments ± standard deviation.
3280 Download full-text
J.S. Nanra et al. / Vaccine 27 (2009) 3276–3280
antigen expression by bacterial isolates and/or the in vivo envi-
ronment encountered by the bacteria. We believe that a vaccine
consisting of more than one antigen has a higher likelihood of suc-
gens in an animal model . In this study, passive immunization
ing bacterial burden in a mouse mastitis model. Our data support
a multiantigen vaccine approach consisting of antigens that are
expressed during different phases of infection in varying animal
The authors would like to thank Dr. S. M. Baker for his insightful
discussions. The S. aureus strains 9596, 9570, 9528, 9576, 9580, and
9934 were kindly provided by Dr. P. Bradford (Wyeth, Pearl River,
NY). CDC3, CDC11, CDC12, CDC13 and CDC17 were kindly provided
by Dr. F. Tenover (CDC, Atlanta, GA). Strain 10938 and UAMS-1 were
obtained from Dr. B. Kreiswirth (UMDNJ, Newark, NJ) and Dr. M.
Smeltzer (UAMS, Little Rock, AK) respectively. We would also like
to thank Dr. J. Patti at Inhibitex for discussion and providing the fol-
 Dryla A, Prustomersky S, Gelbmann D, Hanner M, Bettinger E, Kocsis B,
et al. Comparison of antibody repertoires against Staphylococcus aureus in
healthy individuals and in acutely infected patients. Clin Diagn Lab Immunol
 Verbrugh HA, Peterson PK, Nguyen BY, Sisson SP, Kim Y. Opsonization of encap-
sulated Staphylococcus aureus: the role of specific antibody and complement. J
 Stephan F, Yang K, Tankovic J, Soussy CJ, Dhonneur G, Duvaldestin P, et al.
Impairment of polymorphonuclear neutrophil functions precedes nosocomial
infections in critically ill patients. Crit Care Med 2002;30(2):315–22.
 Cunnion KM, Zhang HM, Frank MM. Availability of complement bound to
Staphylococcus aureus to interact with membrane complement receptors influ-
ences efficiency of phagocytosis. Infect Immun 2003;71(2):656–62.
 Cunnion KM, Benjamin Jr DK, Hester CG, Frank MM. Role of complement recep-
aureus bacteremia. J Lab Clin Med 2004;143(6):358–65.
 Middleton JR. Staphylococcus aureus antigens and challenges in vaccine devel-
opment. Expert Rev Vac 2008;7(6):805–15.
 Thakker M, Park JS, Carey V, Lee JC. Staphylococcus aureus serotype 5 capsular
polysaccharide is antiphagocytic and enhances bacterial virulence in a murine
bacteremia model. Infect Immun 1998;66(11):5183–9.
 Roghmann M, Taylor KL, Gupte A, Zhan M, Johnson JA, Cross A, et al. Epidemiol-
ogy of capsular and surface polysaccharide in Staphylococcus aureus infections
complicated by bacteraemia. J Hosp Infect 2005;59(1):27–32.
 Fattom A, Schneerson R, Watson DC, Karakawa WW, Fitzgerald D, Pastan
I, et al. Laboratory and clinical evaluation of conjugate vaccines com-
posed of Staphylococcus aureus type 5 and type 8 capsular polysaccharides
bound to Pseudomonas aeruginosa recombinant exoprotein A. Infect Immun
 Lee JC, Park JS, Shepherd SE, Carey V, Fattom A. Protective efficacy of antibodies
of endocarditis in rats. Infect Immun 1997;65(10):4146–51.
 Fattom AI, Sarwar J, Ortiz A, Naso RA. Staphylococcus aureus capsular polysac-
charide (CP) vaccine and CP-specific antibodies protect mice against bacterial
challenge. Infect Immun 1996;64(5):1659–65.
 Han HR, Pak II S. Guidry A. Prevalence of capsular polysaccharide (CP) types of
Staphylococcus aureus isolated from bovine mastitic milk and protection of S.
aureus infection in mice with CP vaccine. J Vet Med Sci 2000;62(12):1331–3.
 Foster TJ, Hook M. Surface protein adhesins of Staphylococcus aureus. Trends
 Josefsson E, Higgins J, Foster TJ, Tarkowski A. Fibrinogen binding sites P336 and
Y338 of clumping factor A are crucial for Staphylococcus aureus virulence. PLoS
 Arrecubieta C, Matsunaga I, Asai T, Naka Y, Deng MC, Lowy FD. Vaccination with
clumping factor A and fibronectin binding protein A to prevent Staphylococcus
aureus infection of an aortic patch in mice. J Infect Dis 2008;198(4):571–5.
 Kuklin NA, Clark DJ, Secore S, Cook J, Cope LD, McNeely T, et al. A novel Staphy-
lococcus aureus vaccine: iron surface determinant B induces rapid antibody
responses in rhesus macaques and specific increased survival in a murine S.
aureus sepsis model. Infect Immun 2006;74(4):2215–23.
 Shinefield HR. Use of a conjugate polysaccharide vaccine in the prevention of
invasive staphylococcal disease: is an additional vaccine needed or possible?
Vaccine 2006;24(Suppl. 2:S2):65–9.
 Kiser KB, Cantey-Kiser JM, Lee JC. Development and characterization of
a Staphylococcus aureus nasal colonization model in mice. Infect Immun
 Arbeit RD, Dunn RM. Expression of capsular polysaccharide during experimen-
tal focal infection with Staphylococcus aureus. J Infect Dis 1987;156(6):947–52.
 Lee JC, Takeda S, Livolsi PJ, Paoletti LC. Effects of in vitro and in vivo growth
conditions on expression of type 8 capsular polysaccharide by Staphylococcus
aureus. Infect Immun 1993;61(5):1853–8.
 Hall AE, Domanski PJ, Patel PR, Vernachio JH, Syribeys PJ, Gorovits EL,
et al. Characterization of a protective monoclonal antibody recognizing
Staphylococcus aureus MSCRAMM protein clumping factor A. Infect Immun
 Yokoyama WM, Christensen M, Santos GD, Miller D. Production of monoclonal
antibodies. Current protocols in immunology; 2006 [Chapter 2:Unit 2.5].
 Romero-Steiner S, Libutti D, Pais LB, Dykes J, Anderson P, Whitin JC, et al. Stan-
dardization of an opsonophagocytic assay for the measurement of functional
cells. Clin Diagn Lab Immunol 1997;4(4):415–22.
 Montgomery CP, Boyle-Vavra S, Adem PV, Lee JC, Husain AN, Clasen J, et
al. Comparison of virulence in community-associated methicillin-resistant
Staphylococcus aureus pulsotypes USA300 and USA400 in a rat model of pneu-
monia. J Infect Dis 2008;198(4):561–70.
sion of Staphylococcus epidermidis SdrG increases following exposure to an in
vivo environment. Infect Immun 2008;76(7):2950–7.
 Cook J, Hepler R, Pancari G, Kuklin N, Fan H, Wang XM, et al. Staphylococcus
of staphylococcal infection. Hum Vaccin 2009;5(3).
polysaccharide and clumping factor A prevent mastitis and the emergence of
unencapsulated and small-colony variants of Staphylococcus aureus in mice.
Infect Immun 2008;76(12):5738–44.