Content uploaded by Ugo D'Oro
Author content
All content in this area was uploaded by Ugo D'Oro on Jun 19, 2017
Content may be subject to copyright.
of June 19, 2017.
This information is current as Dendritic Cells
Macrophage Compartments and Follicular
of Unprocessed Antigen in Lymph Node
Vaccine Adjuvant MF59 Promotes Retention
Diego Piccioli
Carmine Malzone, Ennio De Gregorio, Ugo D'Oro and
Rocco Cantisani, Alfredo Pezzicoli, Rossella Cioncada,
http://www.jimmunol.org/content/194/4/1717
doi: 10.4049/jimmunol.1400623
January 2015;
2015; 194:1717-1725; Prepublished online 14J Immunol
Material
Supplementary 3.DCSupplemental
http://www.jimmunol.org/content/suppl/2015/01/14/jimmunol.140062
References http://www.jimmunol.org/content/194/4/1717.full#ref-list-1
, 8 of which you can access for free at: cites 33 articlesThis article
Subscription http://jimmunol.org/subscription is online at: The Journal of ImmunologyInformation about subscribing to
Permissions http://www.aai.org/About/Publications/JI/copyright.html
Submit copyright permission requests at:
Email Alerts http://jimmunol.org/alerts
Receive free email-alerts when new articles cite this article. Sign up at:
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Immunologists, Inc. All rights reserved.
Copyright © 2015 by The American Association of
1451 Rockville Pike, Suite 650, Rockville, MD 20852
The American Association of Immunologists, Inc.,
is published twice each month byThe Journal of Immunology
by guest on June 19, 2017http://www.jimmunol.org/Downloaded from by guest on June 19, 2017http://www.jimmunol.org/Downloaded from
The Journal of Immunology
Vaccine Adjuvant MF59 Promotes Retention of Unprocessed
Antigen in Lymph Node Macrophage Compartments and
Follicular Dendritic Cells
Rocco Cantisani, Alfredo Pezzicoli, Rossella Cioncada, Carmine Malzone,
Ennio De Gregorio, Ugo D’Oro, and Diego Piccioli
Ag retention within lymph nodes (LNs) upon vaccination is critical for the development of adaptive immune responses, because it
facilitates the encounter of the Ag with cognate lymphocytes. During a secondary exposure of the immune system to an Ag, immune
complexes (ICs) that contain the unprocessed Ag are captured by subcapsular sinus macrophages and are transferred onto follicular
dendritic cells, where they persist for weeks, facilitating Ag presentation to cognate memory B cells. The impact of adjuvants on Ag
retention within the draining LNs is unknown. In this article, we provide the first evidence, to our knowledge, that the oil-in-water
emulsion adjuvant MF59 localizes in subcapsular sinus and medullary macrophage compartments of mouse draining LNs, where
it persists for at least 2 wk. In addition, we demonstrate that MF59 promotes accumulation of the unprocessed Ag within these LN
compartments and facilitates the consequent deposition of the IC-trapped Ag onto activated follicular dendritic cells. These
findings correlate with the ability of MF59 to boost germinal center generation and Ag-specific Ab titers. Our data suggest that
the adjuvant effect of MF59 is, at least in part, due to an enhancement of IC-bound Ag retention within the LN and offer insights
to improve the efficacy of new vaccine adjuvants. The Journal of Immunology, 2015, 194: 1717–1725.
Antigen uptake within lymphoid organs is a crucial step for
the development of an immune response (1–3). Follicular
dendritic cells (FDCs) have been described to play a key
role in the induction of B cell responses, particularly during re-
exposure of the immune system to a protein Ag (3–6). FDCs
capture circulating immune complexes (ICs) through the CR2
(CD21/35) and retain them for a long period (up to 3 wk) (3, 5, 7–
9). A prolonged exposure of the ICs on the FDC surface facilitates
the encounter of the unprocessed Ag with cognate memory B cells
(9, 10). Moreover, FDCs secrete chemokines, such as CXCL13, to
attract and interact with cognate B cells (5, 6, 9). Through this
interaction, FDCs provide Ag-specific B cells with signals from
the cognate Ag and the costimulatory surface or soluble mole-
cules, which have to be integrated to achieve the selective acti-
vation of the memory B cells (4–6, 8–11). FDCs maintain intact
and unprocessed ICs, internalizing and recirculating them on the
surface through a nondegradative vesicle trafficking pathway (12).
In addition, it has been demonstrated that cognate B cells acquire
a portion of the FDC membrane together with the Ag (9). B cell
activation determines both the expansion of the Ag-specific
memory B cell compartment and the formation of plasma cells
(13–15). Thus, FDCs exert a primary role in the induction of
a humoral immune response. In addition, it has been discovered
that lymphoid organ dendritic cells (DCs) sample the Ag-loaded
FDCs to acquire Ag for T cell activation, suggesting that FDCs
have a broader function in the immune response and highlighting
their importance (16). Also, the lymphoid organ macrophages
have been described to play a critical role in the induction of the
humoral immune response (3, 17–20). In particular, it has been
observed that the subcapsular sinus macrophages (SCSMs)
transport ICs onto noncognate B cells that shuttle them onto FDCs
(3, 19, 20). At the same time, IC-loaded SCSMs can relay intact
Ag directly to cognate memory B cells, providing the signal for
B cell activation (3, 17–20). These findings unveiled an unex-
pected and peculiar feature of the SCSMs that, differently from
the “conventional” macrophages, display a low degradation rate of
the engulfed Ag (20).
Adjuvants can be added to vaccine formulations, to significantly
enhance the humoral and cellular immune response to a vaccine
Ag, and this may be particularly relevant for subunit vaccines,
which tend to be poorly immunogenic (21–24). The oil-in-water
emulsion MF59 is a very potent and safe adjuvant licensed for
human use in the European Union (21–24). Previous studies have
shown that, upon i.m. immunization, MF59 facilitates Ag uptake
by immune cells (25), stimulates the expression of innate immune
genes at the injection site, resulting in the recruitment of immune
cells within the muscle, and increases the number of Ag-positive
leukocytes in the draining lymph nodes (LNs) (26, 27). It has
also been demonstrated that MF59 promotes the differentiation of
human monocytes to DCs in vitro (25). However, no information
exists on the effect of MF59 or other adjuvants on the Ag distri-
bution within the intact lymphoid organs during vaccination.
Further, most of the studies analyzing Ag deposition in LNs were
performed in passively immunized animals (9, 10, 19, 20).
In this study, we evaluated Ag capture within draining LNs of
mice, during an endogenous immune response to an Ag formulated
with MF59. As a model Ag, we used PE, taking advantage of the
strong intrinsic fluorescence of this protein (28). Because the char-
acteristic fluorescence of PE is sensitive to protein processing and
Novartis Vaccines and Diagnostics, 53100, Siena, Italy
Received for publication March 7, 2014. Accepted for publication December 9, 2014.
This work was supported by Ministero dell’Istruzione, dell’Universita
`e della Ricerca
Grant PON01_00117.
Address correspondence and reprint requests to Dr. Diego Piccioli, Novartis Vaccines
and Diagnostics, via Fiorentina 1, 53100, Siena, Italy. E-mail address: diego.
piccioli@novartis.com
The online version of this article contains supplemental material.
Abbreviations used in this article: DC, dendritic cell; FDC, follicular dendritic cell;
GC, germinal center; IC, immune complex; LN, lymph node; MM, medullary mac-
rophage; SCS, subcapsular sinus; SCSM, subcapsular sinus macrophage.
Copyright Ó2015 by The American Association of Immunologists, Inc. 0022-1767/15/$25.00
www.jimmunol.org/cgi/doi/10.4049/jimmunol.1400623
by guest on June 19, 2017http://www.jimmunol.org/Downloaded from
requires the integrity of PE phycobilosome (28, 29), the use of this
protein allows the detection of intact and unprocessed Ag, which
retains all B cell epitopes and their immunogenic potential for B cell
activation. Although previous studies on Ag deposition have been
conducted upon s.c. or footpad injection of the Ag (10, 15, 19, 20,
30, 31), we decided to use i.m. immunization because this is the
preferred route for vaccine administration in humans.
Materials and Methods
Mice and immunizations
C57BL/6 mice aged 4–6 wk were purchased from Charles River Labora-
tories. Animal experiments were performed in compliance with the Eu-
ropean guidelines and approved by the Novartis internal Animal Welfare
Body. Mice were immunized i.m. (quadriceps) in one leg (unless stated
otherwise) with a volume of 50 ml, using PBS as dilution buffer. PE
(Molecular Probes, Invitrogen Life Technologies) Ag was used at 6 mg/
dose. MF59 was diluted 50% v/v per dose. Immunization schedules are
described in the Results. MF59 labeled with DiD lipophilic tracer (Mo-
lecular Probes, Invitrogen Life Technologies) was used to track the adju-
vant localization within LNs.
Confocal microscopy
LNs of mice were collected under dry conditions at the appropriate time
points, were immediately frozen using liquid nitrogen, and were stored at
280˚C until processing. Cryosections of LNs obtained with the cryostat
CM1950 (Leica) were stained using the following Abs: anti–CD21/35-
Pacific Blue (Biolegend) or -FITC (eBioscience) to identify FDCs; anti–
CD169-FITC (Serotec) to identify SCSMs; anti–F4/80-Alexa Fluor 450 to
identify medullary macrophages (MMs); anti–GL7-FITC (BD Pharmin-
gen) and anti–IgD-biotin (eBioscience) plus streptavidin-Alexa Fluor 405
(Invitrogen Life Technologies) to identify germinal center (GC) B cells;
and anti-CD16/32-Pacific Blue or -PE to identify FDC activation. The
cryosections (8 mm thick) were cut along the entire organ to analyze all the
planes of the organs. The cryosections were fixed using PBS/3% formal-
dehyde for 10 min at room temperature, washed twice with PBS, and
permeabilized with PBS/3% BSA/1% saponin (permeabilization buffer)
for 30 min at room temperature. Tissue sections were then incubated with
the appropriate Abs diluted in permeabilization buffer for 1 h at room
temperature in the dark. An additional incubation step with streptavidin
after washing was performed for IgD staining. After washing three times
with permeabilization buffer and once with PBS, stained tissue sections
were sealed using Gold Antifade reagent (Invitrogen Life Technologies)
and a coverslip. Images were acquired with the Axio Observer LSM700
confocal microscope (Zeiss) at 20˚C, using FLUAR 53or Plan-
Apochromat 403objective lenses with 0.25 or 1.3 of numerical aperture,
respectively. The 403objective lens was used with the Zeiss Immersol
518F imaging medium. Images were processed with Zen 2008 software
(Zeiss).
Flow cytometry
LNs of mice were collected and immediately subjected to enzymatic di-
gestion. In brief, LNs of each group of mice specifically treated were pooled
in RPMI 1640 medium (Life Technologies, Invitrogen Life Technologies)
containing Liberase Research Grade (Roche) at the working concentration
of 500 mg/ml and DNase I (Roche) at the working concentration of 250 mg/
ml (digestion buffer). LNs were incubated for 30 min at 37˚C and then
agitated by pipetting every 15 min. Supernatant was collected and fresh
digestion buffer was added. The procedure was repeated three times. When
the LNs were totally disintegrated, all the supernatants were pooled and the
total LN cells were collected by centrifugation at 300 3gfor 10 min, at
room temperature. Then the LN cells were washed with RPMI 1640 me-
dium (repeating the centrifugation) and were suspended in RPMI 1640
medium supplemented with 10% FCS (HyClone), 1% PSG (EuroClone).
Cells were filtered with a 70-mm cell strainer (Falcon, Becton Dickinson)
and counted with a hematocytometer. Total LN cells were labeled with
live/dead yellow (Invitrogen Life Technologies), anti–CD169-Alexa Fluor
488 (Serotec), anti–F4/80-Pacific Blue (eBioscience), anti–CD11c-allo-
phycocyanin (Becton Dickinson), and anti–CD19-allophycocyanin-Cy7
(Pharmingen, Becton Dickinson). In brief, LN cells were incubated with
live/dead staining and Abs for 20 min at 4˚C in dark conditions. Cells were
washed with PBS by centrifugation at 300 3gfor 10 min at room tem-
perature. Washed cells were suspended in PBS and analyzed by flow
cytometry using an LSR II cytometer (Becton Dickinson).
Detection of anti-PE Abs
Serum PE-specific total IgG were measured by ELISA. In brief, Maxisorb
96-well plates (Nunc) were coated with a PE solution (2.5 mg/ml) in
carbonate buffer (100 ml/well) overnight at 4˚C. Plates were then blocked
by addition of PBS 3%/polyvinylpyrrolidone (SERVA) (200 ml/well), in-
cubated for 2 h at 37˚C, and then washed once with PBS, 0.05%/Tween 20
(washing buffer). Serial dilutions (3-fold step) of standard and serum
samples in PBS/0.05% Tween 20/1% BSA were added to the wells and
incubated for 2 h at 37˚C. Plates were then washed three times with
washing buffer and incubated for 1 h at 37˚C with anti-mouse IgG-alkaline
phosphatase (Sigma-Aldrich) solution (100 ml/well). After three washes,
the substrate p-nitrophenylphosphate (100 ml/well; Sigma-Aldrich) was
added for 30 min at room temperature. Absorbance to 405 nm was then
measured by a plate spectrophotometer (BioTek-ASHI).
Results
Adjuvant MF59 induces Ag retention within draining LNs
We sought to determine whether the adjuvant MF59 affects Ag
uptake and distribution within LNs. To analyze the distribution of
the injected Ag in the presence of naturally formed ICs, we im-
munized mice twice (14 d apart) with PE in the presence or absence
of MF59 and collected draining LNs and sera of mice 1 h, 6 h, and
7 d after the second immunization (Fig. 1). After just 1 h, the Ag
was already detectable within the subcapsular sinus (SCS) and
medullary compartments of the LN in MF59-treated mice, as
shown by colocalization with CD169 (SCSM) and F4/80 (MM),
whereas no Ag was detected in mice immunized with PE alone
(Fig. 1A, 1F, 1G). The PE Ag was undetectable after the primary
immunization either at early or at late time points, such as 14 d
upon priming, which is the booster time point (data not shown).
As previously established in passively immunized mice, 6 h after
the second immunization we found some of the Ag colocalized
with FDCs in mice immunized with MF59-adjuvanted PE,
whereas no Ag was detectable in the absence of adjuvant (Fig. 1B,
1F, 1G). Surprisingly, at this time point, most of the Ag was still
present in the area of the LN corresponding to SCS and medullary
compartments (Fig. 1B, 1F, 1G), indicating a persistence of the Ag
in the macrophage compartments, in the presence of MF59. Fi-
nally, as expected, 7 d after the second immunization with MF59,
the Ag was totally retained by FDCs, as indicated by colocaliza-
tion with the CD21/35 marker, whereas no PE was detectable onto
FDCs in absence of MF59 (Fig. 1C, 1F). In addition, MF59 in-
duced a marked generation of GCs, as detected by GL7 staining
(Fig. 1C), which is consistent with a greater production of anti-PE
Abs compared with immunization in the absence of adjuvant
(Fig. 1D). Complement activation and binding of ICs to CD21/35
through the C3d complement fragment are both required for Ag
deposition onto FDCs in passively immunized mice and for the
generation of an optimal humoral immune response (8). To verify
whether complement activation was necessary for PE deposition
onto FDCs also in our experimental setting, we used C3
2/2
mice,
in which the formation of ICs containing C3 fragments is abol-
ished. When these mice were immunized with PE in the presence
or absence of MF59, we observed an impaired Ag deposition onto
FDCs and, more importantly, an impaired GC formation, even in
the presence of the adjuvant (Fig. 1E, 1F). This finding, which is
in agreement with published data, demonstrates that the increased
IC-bound Ag deposition onto FDCs observed in the presence of
MF59 is dependent on complement activation and is required for
an optimal B cell response.
To quantify the Ag retention within draining LNs in our ex-
perimental setting, we analyzed the LNs by flow cytometry (Fig. 2).
Because 6 h after the second immunization is the key time point
for Ag retention in our experimental setting, we focused our at-
tention on this time point. Quantitation of PE-loaded cells by flow
1718 ADJUVANT MF59 PROMOTES Ag ACCUMULATION WITHIN LNs
by guest on June 19, 2017http://www.jimmunol.org/Downloaded from
cytometry recapitulates and extends what was observed by con-
focal microscopy (Fig. 2). We found a notable increase in the
number and percentage of PE
+
total LN cells in mice immunized
in the presence of MF59 (Fig. 2A), and this result was observed
FIGURE 1. MF59 adjuvant induces Ag retention within SCS and medullary compartments of draining LNs. (A–Cand E) Wilde type (A–C)orC3
2/2
(E)
C57BL/6 mice were immunized twice i.m. with PE (upper panels) and PE + MF59 (lower panels). Inguinal draining LNs were collected 1 h (A),6h(B), or
7d(Cand E) after the second immunization and analyzed by confocal microscopy to detect SCSMs (CD169-FITC) and MMs (F4/80-Alexa Fluor 450) (A)
or FDCs (CD21/35-Pacific Blue) and GC B cells (GL7-FITC) (B,C, and E). Signals that localize CD169, F4/80, CD21/35, or GL7 expression and PE are
shown separately or merged as indicated. The image of one section is shown, in each panel, as an example of consecutive sections of a whole LN, which is
representative of three organs from different mice. Original magnification 35. Scale bar, 100 mm. One representative experiment of two is shown. (D) Anti-
PE IgG titers in the sera of the same mice that LNs are shown in (B) (6 h after the second immunization) and (C) (7 d after the second immunization). Each
symbol represents the serum anti-PE IgG titer of a single mouse. Anti-PE IgG titers were measured by ELISA. **p,0.01, unpaired Student ttest. (F)
Schematic representation of a LN to facilitate the interpretation of the confocal microscopy images. (G) Typical confocal microscopy detection of MMs
(F4/80) and SCSMs (CD169) of inguinal LN from naive untreated mice. Signals that localize F4/80 or CD169 expression are shown separately or merged as
indicated. The image of one section is shown, as an example of consecutive sections of a whole LN, which is representative of several organs from different
mice. Original magnification 35. Scale bar, 100 mm.
The Journal of Immunology 1719
by guest on June 19, 2017http://www.jimmunol.org/Downloaded from
for all APC subsets, including not only SCSMs and MMs, but also
DCs and B cells (Fig. 2B). Unlike confocal microscopy, by using
flow cytometry, we were able to detect PE within the draining LNs
even in the absence of MF59 coadministration (Fig. 2). This dif-
ference was presumably due to the greater sensitivity of flow
cytometry compared with confocal microscopy. Interestingly, fo-
cusing on PE
+
cells, we observed that, on immunization with
MF59, most PE-loaded cells are B cells (Fig. 2C). However, the
Ag loading by macrophages, revealed as PE fluorescence intensity,
is much higher than the Ag loading by B cells (Fig. 2D). Similar
results were obtained also at 1 h after the boost, albeit to a lesser
extent (data not shown). Taken together, these findings are con-
FIGURE 2. Quantitation of Ag retention induced by MF59 within draining LNs. Similar experiments of Fig. 1. Inguinal draining LNs were collected 6 h
after the second immunization and analyzed by flow cytometry. (A) Number (left panel) or percentage (right panel)ofPE
+
cells within total draining LN
cells from mice immunized with PE alone (black bars) or PE adjuvanted with MF59 (white bars). (B) Number (left panel) or percentage (right panel)ofPE
+
SCSMs (phenotypically distinguished as CD169
+
-F4/80
2
-CD11c
2
), MMs (phenotypically distinguished as CD169
+
-F4/80
+
CD11c
2
), DCs (phenotypically
distinguished as CD169
2to+
-F4/80
2to+
-CD11c
+
) and B cells (phenotypically distinguished as CD19
+
) as indicated, within total draining LN cells from mice
immunized with PE alone (black bars) or PE adjuvanted with MF59 (white bars). (C) Percentage of PE
+
B cells among the PE
+
total LN cells from mice
immunized with PE alone (black bar) or adjuvanted with MF59 (white bar). (D) Comparison of PE loading by B cells (green bars), MMs (blue bars), and
SCSMs (red bars) in immunization with MF59. The mean fluorescence intensity (MFI) of PE for each cell population is depicted as the difference (DMFI;
left panel) or the ratio (MFI fold increase; right panel) with the background (PBS-injected mice). The DMFI reveals that the net PE fluorescence of
macrophages is roughly one logarithmic higher than the net PE fluorescence of B cells, whereas the MFI fold increase confirms that the PE uptake by
macrophages is higher than the PE uptake by B cells. Inguinal draining LNs from three different mice per treatment were pooled. Results from one
representative experiment of two are depicted.
1720 ADJUVANT MF59 PROMOTES Ag ACCUMULATION WITHIN LNs
by guest on June 19, 2017http://www.jimmunol.org/Downloaded from
sistent with the confocal microscopy observation and confirm the
IC-bound Ag translocation model from macrophages to non-
cognate B cells toward FDCs. Thus, we demonstrated that MF59
adjuvant enhances Ag accumulation by macrophages and conse-
quently by noncognate B cells, over the first 6 h, ultimately
leading to Ag retention onto FDCs. In contrast with the confocal
microscopy data, at 7 d after the boost, we were able to detect very
low amounts of PE
+
cells within the total LN cells (data not shown).
This expected result is likely due to the fact that the enzymatic
digestion of LNs induces the loss of stromal cells, including FDCs,
wherealltheIC-trappedAgislocated7daftertheboostasshown
in Fig. 1 and as reported in the literature (1–3, 5, 6, 9).
FIGURE 3. MF59 directly affects Ag retention
within SCS and medullary compartments. (A–C) Mice
were immunized i.m. in both legs with PE alone (B)or
PE + MF59 (C), and serum anti-PE IgG titers were
measured by ELISA after 2 wk (A). (A) Anti-PE IgG
titers. Each symbol represents the serum anti-PE IgG
titer of a single mouse. Unpaired Student ttest does not
reveal any statistical significant difference between the
two groups. (Band C) Each group of mice was suc-
cessively reimmunized in the left leg with PE alone
(upper panels) and in the right leg with MF59-adju-
vanted PE (PE + MF59, lower panels). Then 6 h after
the second immunization, inguinal draining LNs were
collected and analyzed by confocal microscopy to de-
tect FDCs (CD21/35-Pacific Blue), SCSMs (CD169-
FITC), and PE (Band C). Signals that localize CD21/35
or CD169 expression and PE are shown separately or
merged as indicated. The image of one section is
shown, in each panel, as example of consecutive sec-
tions of a whole LN, which is representative of three
(B) or two (C) organs from three different mice. Orig-
inal magnification 35. Scale bar, 100 mm. (C)Bottom
panels 1 and 2are enlargements of the areas indicated
as 1and 2in the Merge panel above. One representative
experiment of two is shown.
The Journal of Immunology 1721
by guest on June 19, 2017http://www.jimmunol.org/Downloaded from
Adjuvant MF59 directly affects Ag retention within draining LNs
Detection of the Ag in the macrophage area after a booster im-
munization in the presence of MF59 could theoretically be a
consequence of a superior primary Ab response and an increased
amount of circulating ICs, because of the MF59 adjuvant effect
during priming immunization. Indeed, the anti-PE Ab titer 6 h after
the second immunization is higher in the presence of MF59, al-
though this difference is not statistically significant (Fig. 1D). We
therefore designed an experiment to better evaluate the Ag re-
tention within LN macrophage compartments after secondary
immunization with or without MF59, but in the presence of an
identical amount of circulating ICs. We first immunized mice in
both legs with PE alone or adjuvanted with MF59; then for the
secondary immunization (2 wk later), the animals of both groups
were injected with PE alone in the left leg and with MF59-
adjuvanted PE in the right leg, and the draining LNs of both
legs were analyzed. In animals treated with PE alone as the first
immunization and thus in the presence of very low amounts of Ag-
specific Abs (and consequently ICs) (Fig. 3A), we did not detect
Ag in SCS or medullary compartment 6 h after the secondary
injection with either plain or adjuvanted PE (Fig. 3B). Thus, the
secondary immunization with MF59 is not sufficient to accumu-
late within the draining LN enough Ag to be detected after
priming with PE alone (in the presence of low amounts of Ag-
specific Abs). On the contrary, in animals that received MF59-
adjuvanted PE as the primary immunization and that displayed
higher (but not statistically significant) anti-PE Ab titers (Fig. 3A),
MF59 was required at the secondary immunization to promote an
optimal retention of PE within the SCS and medullary compart-
ments in the draining LNs (Fig. 3C). Indeed, the second immu-
nization with PE resulted in an efficient Ag retention within the
draining LN of the leg treated with two immunizations in the
presence of MF59 (Fig. 3C). On the contrary, in the same mouse,
within the draining LN of the leg that received the primary im-
munization with MF59-adjuvanted PE and the boost with PE
alone, we did not observe an efficient Ag retention (Fig. 3C), and
in only one mouse of three could we detect some PE signal within
the macrophage compartments (Supplemental Fig. 1). Thus, ad-
juvant MF59 play a significant role for an optimal LN retention
of the IC-bound Ag during the secondary immunization in the
presence of determined amounts of Ag-specific Abs. These results
also confirm that the primary immunization with MF59 does not
promote enough Ag accumulation to be detectable at the booster
time point. The IC-bound Ag was clearly detectable in the mac-
rophage area and, as expected, in the process of translocation
toward FDCs (enlarged images 1 and 2 at bottom of Fig. 3C). Ag
retention by macrophages precedes its translocation inside the
B cell follicles, and this step allows the subsequent deposition of
the Ag onto FDCs. Thus, MF59 directly stimulates IC-bound Ag
accumulation within LNs, even if the contribution of circulating
ICs could be critical over a threshold amount.
To further assess the direct contribution of MF59 in the retention
of the Ag within macrophage compartments, we evaluated whether
MF59 reaches draining LNs upon injection. Indeed, labeled MF59
waspresentintheSCSandmedullarycompartments1and6hafter
injection (Fig. 4A, 4B) and remained in these compartments for a
long time, being detectable after 7 (Fig. 4C) and 14 d (Fig. 4D). In
addition, at day 7, some MF59 was also detected within the para-
cortex of the LN, which is the T cell area where DCs reside (Fig. 4C
and its magnification). Therefore, MF59 never reaches the FDC area
or the B cell follicles, but remains mainly within the macrophage area.
Upon immunizations with MF59, FDC activation is associated
with Ag trapping and GC generation
Given the importance of FDCs for the development of a humoral
immune response, we asked whether the MF59 adjuvant could
affect FDC activation and whether this activation correlates with
generation of GCs and Ag uptake by FDCs. We first analyzed
activation of FDCs in the LNs of mice immunized with PE in the
presence or absence of MF59, 7 d after a secondary immunization,
checking the expression of CD16/32, a described marker for FDC
activation (7, 32). The activation of FDCs was generally more
evident after immunization with MF59, although it was also ob-
served in the absence of adjuvant (Fig. 5A). However, in mice
immunized with MF59-adjuvanted PE, Ag deposition onto FDCs
was observed only in follicles displaying activated FDCs and
FIGURE 4. MF59 is retained within SCS and medullary compartments. Mice were injected i.m. with DiD-labeled MF59, and inguinal draining LNs were
collected after 1 h (A),6h(B),7d(C), and 14 d (D) and analyzed by confocal microscopy to detect SCSMs (CD169-FITC), MMs (F4/80-Alexa Fluor 450),
FDCs (CD21/35-Pacific Blue), and labeled MF59. Signals that localize CD169, F4/80, or CD21/35 expression and MF59 are shown separately or merged as
indicated. (C,bottom panel) Magnification of the indicated area in the Merge image. The image of one section is shown, in each panel, as an example of
consecutive sections of a whole LN, which is representative of three organs from different mice. Original magnification 35. Scale bar, 100 mm. One
representative experiment of two is shown.
1722 ADJUVANT MF59 PROMOTES Ag ACCUMULATION WITHIN LNs
by guest on June 19, 2017http://www.jimmunol.org/Downloaded from
matched with CD16/32-expressing cells (Fig. 5B, upper and
central panels), whereas CD16/32 expression could be detected
also in the absence of Ag uptake by FDCs (Fig. 5B, lower panel).
Using a nonfluorescent Ag (RrgB protein from the S. pneumoniae
pilus), we confirmed that FDC activation within draining LNs,
measured detecting the CD16/32 expression, also occurred in the
absence of adjuvant (data not shown). More importantly, in this
experimental system, we found that GC formation correlates with
activation of FDCs. In fact, GCs could be observed only in the
presence of FDC activation (Supplemental Fig. 2), even though
B cell follicles without detectable GC reaction, but with CD16/32
expression in the FDC area, were observed (data not shown).
These results suggest that after immunizations with adjuvant
MF59, FDC activation is necessary, but not sufficient, for Ag trapping
and GC generation.
Discussion
Ag trapping onto FDCs or by SCSMs was previously assessed in
passively immunized mice, to artificially induce IC formation,
using s.c. (flank, tail) or footpad as route of Ag administration (9,
10, 19, 20). In these models, it has been demonstrated that the Ag
reaches the SCS in minutes and within a few hours (6–24 h) is
almost completely deposited in the FDC area, where it can be
trapped for days and can be readily detectable at 1 wk, when GCs
can be observed (9, 10, 15, 19, 20, 30, 31, 33). All these previous
studies have been performed administrating a soluble Ag in mice
artificially reconstituted with circulating Ag-specific Abs, whereas
no study to date has evaluated the effect of vaccine adjuvants on
Ag deposition on the macrophage and FDC area of draining LNs,
in the context of an endogenous immune response.
We decided to investigate whether the vaccine adjuvant MF59
is able to modulate Ag distribution within LNs upon a booster
immunization, by affecting trafficking of an IC-trapped Ag. The
analysis of the kinetics of Ag translocation within the draining LNs
during an endogenous response to an MF59-adjuvanted vaccine
enables a better understanding of the mechanism of action of this
adjuvant. Clarifying this mechanism is highly important because
MF59 is currently used to formulate seasonal and pandemic in-
fluenza vaccines used in humans and has the potential to greatly
improve the efficacy of poorly immunogenic vaccine Ags. All
experiments in our study were performed using i.m. immunization,
because this is the preferred route of administration for human
vaccines. We chose the PE as model Ag because of its strong
intrinsic fluorescent property and its rather poor immunogenicity,
FIGURE 5. Upon immunizations with MF59, FDC
activation is associated with Ag trapping. (Aand B)
Confocal microscopy images of the same LNs of
Fig. 1C analyzed to detect FDCs (CD21/35-FITC) or
FcgRII/III expression (CD16/32-Pacific Blue) and PE.
(A) Images of an entire draining LN from mice im-
munized with PE (upper panel) or PE + MF59 (lower
panel). (B) Magnified images of representative follicles
from LNs depicted in (A,lower panel) display colo-
calization of activated FDCs and PE (upper panel),
nonactivated FDCs in the absence of PE (central
panel), and activated FDCs in the absence of PE (lower
panel). Signals that localize the expression of CD21/
35, or CD16/32 and PE are shown separately or
merged as indicated. The image of one section is
shown, in each panel, as example of consecutive sec-
tions of a whole LN, which is representative of three
organs from different mice. Original magnification 35
(A), 340 (B). Scale bar, 100 mm.
The Journal of Immunology 1723
by guest on June 19, 2017http://www.jimmunol.org/Downloaded from
which allowed us to better evaluate the effect of an adjuvant on the
enhancement of a specific immune response.
Consistently with the described model for the trafficking of
IC-trapped Ags within the LN of passively immunized mice (1, 2,
10, 17–20), we observed that PE formulated with MF59, drained
to the LN from the injection site, first encounters SCSMs and is
then translocated onto FDCs. Interestingly, however, we discov-
ered that MF59 promotes accumulation within the SCSM and MM
compartments, of the unprocessed Ag trapped in ICs, enhancing
its subsequent deposition onto noncognate B cells and then FDCs.
Indeed, no Ag translocation onto FDCs was detected in the ab-
sence of complement C3 factor and, therefore, in the absence of
IC formation. In addition, the Ag accumulation observed after
immunization with MF59 correlates with GC formation and the
increase in the magnitude of the humoral immune response, which
is consistent with the ability of SCSMs and FDCs to deliver Ag to
cognate memory B cells, resulting in their activation. Thus, our
results confirm a correlation between retention of the intact IC-
bound Ag within LNs and development of an optimal humoral
immune response. The novel finding that MF59 adjuvant exerts
a direct effect on this process remaining mainly localized in the
macrophage compartments highlights the importance of this cel-
lular population in adjuvant activity and identifies macrophages as
a critical target of MF59 function. Taken together, our data rep-
resent the first demonstration that, during a secondary immune
response after re-exposure to an Ag upon i.m. immunization, the
presence of the vaccine adjuvant MF59 directly induces retention
for several hours of the intact, unprocessed Ag, trapped in ICs, by
the compartment of SCSMs and MMs. Interestingly, noncognate
B cells play a central role in Ag accumulation within the SCS
and medullary compartments, either confirming the translocation
model of IC-bound Ag from macrophages to B cells or demon-
strating that MF59 enhances the accumulation of the Ag by all the
cells involved in the process of Ag translocation onto FDCs. In
addition, it is conceivable that MF59 induces the recruitment of
noncognate B cells within the LN macrophage compartments to
potentiate the mechanism of translocation onto FDCs. Moreover,
the accumulation of unprocessed IC-bound Ag in the SCS and
medullary compartments is followed by its strong deposition onto
FDCs, which correlates with a robust humoral response.
The exact mechanism by which MF59 exerts this effect on Ag
accumulation is still poorly understood. MF59 might affect the
degradation of IC-bound Ag by APCs once engulfed, but this
hypothesis has not been addressed. However, it was previously
proposed that peripheral macrophages, loaded with Ag and MF59,
promote Ag translocation into LNs (27); further experiments are
needed to confirm this hypothesis. Certainly, it has been demon-
strated that MF59 induces inflammation at injection site and
increases the Ag-positive leukocytes within the draining LNs (26,
27). Thus, the migration of immune cells from injection site into
the draining LN could play a critical role in the enhancement of
Ag and MF59 accumulation within the node. In conclusion, al-
though the details of these phenomena are still to be elucidated,
our study has provided the first demonstration, to our knowledge,
that MF59 promotes IC-bound Ag retention within draining LNs
during the booster response.
FDCs are critical for the induction of a complete humoral im-
mune response, because these cells are required to capture IC-bound
Ags and stimulate cognate B cells (9, 10), leading to an increase of
the Ag-specific memory B cell population and expansion of the
Ab-secreting plasma cells (13–15). FDCs have also been shown to
function as a reservoir of Ag for uptake by lymphonodal DCs (16).
However, no study has investigated FDC activation, in vivo, after
vaccination. We discovered that FDCs undergo activation upon
immunization with a poorly immunogenic Ag, such as PE, even in
the absence of an adjuvant, and that FDC activation is increased by
addition of MF59. Therefore, we propose that, although a weak
immune response can induce FDC activation, the presence of ad-
juvant MF59 in the immunization can modulate the function of
FDCs. Moreover, our study provides the first demonstration that,
upon vaccination, the activation of FDCs is necessary for both GC
generation and Ag deposition. In fact, consistent with the ability of
FDCs to be activated after immunization with a poorly immuno-
genic Ag, we observed LN follicles displaying activated FDCs
without detectable GCs or Ag deposition. However, in the same
organ, we also observed GC formation and Ag deposition only in
follicles containing activated FDCs. In addition, the deposited Ag
is clearly colocalized with the activated cells of the FDC network,
indicated by the expression of CD16/32. In a study of passively
immunized mice, it has previously been demonstrated that ex-
pression of CD21/35 is required for Ag deposition onto FDCs,
whereas CD16/32 is dispensable (8). Therefore, based on this
finding and on our data, we suggest that CD16/32 represents
simply an activation marker for FDCs, and that, upon vaccination,
the activation of FDCs is required, but not sufficient, for Ag re-
tention by the cells and GC formation within a single B cell fol-
licle. Interestingly, MF59 may partially modulate this process.
IC-trapped Ag retention by LN macrophages facilitates Ag
translocation to FDCs and Ag uptake by cognate B cells or DCs,
which are key steps in the initiation of an immune response. It is
therefore important to evaluate how this process is modulated by
vaccine adjuvants (9, 10, 16–20). Our study demonstrates that
vaccine adjuvant MF59 promotes retention of the intact, unpro-
cessed Ag trapped in ICs within the LN macrophage compart-
ments, affecting the cascade of events leading to a marked
deposition of this IC-bound Ag onto activated FDCs. This increase
of the overall accumulation of unprocessed Ag within the LNs is
associated with the ability of MF59 to enhance the humoral im-
mune response. In conclusion, our work identifies a new mecha-
nism by which MF59 may increase vaccine immunogenicity and
will help to improve the efficacy of new vaccine adjuvants.
Acknowledgments
We thank Marco Tortoli and Elena Amantini for performing injection and
collections of sera and LNs, Luis Brito and Dinorah Jean-Gilles for provid-
ing DiD-labeled MF59, Simone Vecchi, Donatello Laera, and Francesco
Doro for providing MF59 formulations, Elena Cartocci and Claudia
Facciotti for purifying RrgB protein from S. pneumoniae pilus, Sydney Lavoie
and Michael C. Carroll for providing C3
2/2
mice, Matthew Bottomley for
linguistic assistance on the manuscript, Marco Soriani for advice on confocal
microscopy experiments, and Matthew Woodruff, Stefano Sammicheli, and
Matteo Iannacone for critical review of the manuscript.
Disclosures
All authors are employees or managers of Novartis Vaccines. MF59 is
a proprietary adjuvant of Novartis.
References
1. Gonzalez, S. F., M. P. Kuligowski, L. A. Pitcher, R. Roozendaal, and
M. C. Carroll. 2010. The role of innate immunity in B cell acquisition of antigen
within LNs. Adv. Immunol. 106: 1–19.
2. Gonzalez, S. F., S. E. Degn, L. A. Pitcher, M. Woodruff, B. A. Heesters, and
M. C. Carroll. 2011. Trafficking of B cell antigen in lymph nodes. Annu. Rev.
Immunol. 29: 215–233.
3. Cyster, J. G. 2010. B cell follicles and antigen encounters of the third kind. Nat.
Immunol. 11: 989–996.
4. Park, C. S., and Y. S. Choi. 2005. How do follicular dendritic cells interact in-
timately with B cells in the germinal centre? Immunology 114: 2–10.
5. Allen, C. D., and J. G. Cyster. 2008. Follicular dendritic cell networks of primary
follicles and germinal centers: phenotype and function. Semin. Immunol. 20: 14–25.
6. Aguzzi, A., and N. J. Krautler. 2010. Characterizing follicular dendritic cells:
a progress report. Eur. J. Immunol. 40: 2134–2138.
1724 ADJUVANT MF59 PROMOTES Ag ACCUMULATION WITHIN LNs
by guest on June 19, 2017http://www.jimmunol.org/Downloaded from
7. El Shikh, M. E., R. El Sayed, A. K. Szakal, and J. G. Tew. 2006. Follicular
dendritic cell (FDC)-FcgammaRIIB engagement via immune complexes induces
the activated FDC phenotype associated with secondary follicle development.
Eur. J. Immunol. 36: 2715–2724.
8. Roozendaal, R., and M. C. Carroll. 2007. Complement receptors CD21 and
CD35 in humoral immunity. Immunol. Rev. 219: 157–166.
9. Suzuki, K., I. Grigorova, T. G. Phan, L. M. Kelly, and J. G. Cyster. 2009. Vi-
sualizing B cell capture of cognate antigen from follicular dendritic cells. J. Exp.
Med. 206: 1485–1493.
10. Roozendaal, R., T. R. Mempel, L. A. Pitcher, S. F. Gonzalez, A. Verschoor,
R. E. Mebius, U. H. von Andrian, and M. C. Carroll. 2009. Conduits mediate
transport of low-molecular-weight antigen to lymph node follicles. Immunity 30:
264–276.
11. Wang, X., B. Cho, K. Suzuki, Y. Xu, J. A. Green, J. An, and J. G. Cyster. 2011.
Follicular dendritic cells help establish follicle identity and promote B cell re-
tention in germinal centers. J. Exp. Med. 208: 2497–2510.
12. Heesters, B. A., P. Chatterjee, Y. A. Kim, S. F. Gonzalez, M. P. Kuligowski,
T. Kirchhausen, and M. C. Carroll. 2013. Endocytosis and recycling of immune
complexes by follicular dendritic cells enhances B cell antigen binding and
activation. Immunity 38: 1164–1175.
13. Wu, Y., S. Sukumar, M. E. El Shikh, A. M. Best, A. K. Szakal, and J. G. Tew.
2008. Immune complex-bearing follicular dendritic cells deliver a late antigenic
signal that promotes somatic hypermutation. J. Immunol. 180: 281–290.
14. Wu, Y., M. E. El Shikh, R. M. El Sayed, A. M. Best, A. K. Szakal, and J. G. Tew.
2009. IL-6 produced by immune complex-activated follicular dendritic cells
promotes germinal center reactions, IgG responses and somatic hypermutation.
Int. Immunol. 21: 745–756.
15. Shapiro-Shelef, M., and K. Calame. 2005. Regulation of plasma-cell develop-
ment. Nat. Rev. Immunol. 5: 230–242.
16. McCloskey, M. L., M. A. Curotto de Lafaille, M. C. Carroll, and A. Erlebacher.
2011. Acquisition and presentation of follicular dendritic cell-bound antigen by
lymph node-resident dendritic cells. J. Exp. Med. 208: 135–148.
17. Carrasco, Y. R., and F. D. Batista. 2007. B cells acquire particulate antigen in
a macrophage-rich area at the boundary between the follicle and the subcapsular
sinus of the lymph node. Immunity 27: 160–171.
18. Pape, K. A., D. M. Catron, A. A. Itano, and M. K. Jenkins. 2007. The humoral
immune response is initiated in lymph nodes by B cells that acquire soluble
antigen directly in the follicles. Immunity 26: 491–502.
19. Phan, T. G., I. Grigorova, T. Okada, and J. G. Cyster. 2007. Subcapsular en-
counter and complement-dependent transport of immune complexes by lymph
node B cells. Nat. Immunol. 8: 992–1000.
20. Phan, T. G., J. A. Green, E. E. Gray, Y. Xu, and J. G. Cyster. 2009. Immune
complex relay by subcapsular sinus macrophages and noncognate B cells drives
antibody affinity maturation. Nat. Immunol. 10: 786–793.
21. McKee, A. S., M. W. Munks, and P. Marrack. 2007. How do adjuvants work?
Important considerations for new generation adjuvants. Immunity 27: 687–690.
22. De Gregorio, E., U. D’Oro, and A. Wack. 2009. Immunology of TLR-
independent vaccine adjuvants. Curr. Opin. Immunol. 21: 339–345.
23. Coffman, R. L., A. Sher, and R. A. Seder. 2010. Vaccine adjuvants: putting
innate immunity to work. Immunity 33: 492–503.
24. Dormitzer,P.R.,G.Galli,F.Castellino,H.Golding,S.Khurana,G.DelGiudice,and
R. Rappuoli. 2011. Influenza vaccine immunology. Immunol. Rev. 239: 167–177.
25. Seubert, A., E. Monaci, M. Pizza, D. T. O’Hagan, and A. Wack. 2008. The
adjuvants aluminum hydroxide and MF59 induce monocyte and granulocyte
chemoattractants and enhance monocyte differentiation toward dendritic cells. J.
Immunol. 180: 5402–5412.
26. Mosca, F., E. Tritto, A. Muzzi, E. Monaci, F. Bagnoli, C. Iavarone, D. O’Hagan,
R. Rappuoli, and E. De Gregorio. 2008. Molecular and cellular signatures of
human vaccine adjuvants. Proc. Natl. Acad. Sci. USA 105: 10501–10506.
27. Calabro, S., M. Tortoli, B. C. Baudner, A. Pacitto, M. Cortese, D. T. O’Hagan,
E. De Gregorio, A. Seubert, and A. Wack. 2011. Vaccine adjuvants alum and
MF59 induce rapid recruitment of neutrophils and monocytes that participate in
antigen transport to draining lymph nodes. Vaccine 29: 1812–1823.
28. Ocarra, P., C. Oheocha, and D. M. Carrroll. 1964. Spectral properties of the
phycobilins. II. Phycoerythrobilin. Biochemistry 3: 1343–1350.
29. Glazer, A. N. 1982. Phycobilisomes: structure and dynamics. Annu. Rev.
Microbiol. 36: 173–198.
30. Fairfax, K. A., A. Kallies, S. L. Nutt, and D. M. Tarlinton. 2008. Plasma cell
development: from B-cell subsets to long-term survival niches. Semin. Immunol.
20: 49–58.
31. Fooksman, D. R., T. A. Schwickert, G. D. Victora, M. L. Dustin,
M. C. Nussenzweig, and D. Skokos. 2010. Development and migration of
plasma cells in the mouse lymph node. Immunity 33: 118–127.
32. El Shikh, M. E., R. M. El Sayed, Y. Wu, A. K. Szakal, and J. G. Tew. 2007.
TLR4 on follicular dendritic cells: an activation pathway that promotes acces-
sory activity. J. Immunol. 179: 4444–4450.
33. Victora, G. D., T. A. Schwickert, D. R. Fooksman, A. O. Kamphorst, M. Meyer-
Hermann, M. L. Dustin, and M. C. Nussenzweig. 2010. Germinal center dy-
namics revealed by multiphoton microscopy with a photoactivatable fluorescent
reporter. Cell 143: 592–605.
The Journal of Immunology 1725
by guest on June 19, 2017http://www.jimmunol.org/Downloaded from