| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Fox Chase Cancer Center, Philadelphia, PA 19111
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) are also good candidates since they
share many of the characteristics of DC including common precursors,
expression of costimulatory molecules, and the ability to move to and
from sites of inflammation. In addition, they can reconstitute CTL
responses in vivo (16).
Because pAPC play such an important role in the orchestration of
antiviral CD8+ T cell immunity, it becomes
important to understand the mechanisms whereby these cells present
viral Ags. As with all other cells, infected pAPC can present
endogenously synthesized viral Ags in vitro and in vivo
(17). In this way, they can initiate responses by a
process known as direct presentation. However, pAPC also have the
unique ability to acquire Ag from exogenous sources and present them on
their own MHC class I molecules. This ability of pAPC to present
exogenous Ags can be readily observed in vitro by coculturing DC or
M with many different antigenic formulations, and also in vivo upon
inoculation of mice with those same Ags (18).
In this paper, we use the term "cross-presentation" to refer
exclusively to a subtype of exogenous Ag presentation that involves the
transfer of Ag from an Ag donor cell (ADC) to pAPC. The important
physiological function of cross-presentation in viral infections became
clear when we demonstrated that cross-presentation is necessary and
sufficient to initiate CTL responses to viruses that do not infect pAPC
(3). Moreover, it is also becoming apparent that
cross-presentation is required to induce CTL responses to viruses that
suppress MHC class I Ag presentation on infected cells (19, 20). In addition, it is very likely that, concurrent with direct
presentation (17), cross-presentation contributes somewhat
to the induction of CTL toward most other viruses. Significantly, it
has been shown that cultured DC phagocytose apoptotic or necrotic cells
that had been infected with influenza virus or irradiated cells that
had been infected with recombinant vaccinia virus and cross-present
viral Ags supplied by these cells (21, 22, 23). However, in
those same reports, M phagocytosed the apoptotic/necrotic cells, but
were unable to cross-present the associated Ags. This led the authors
to propose that cross-presentation of cell-associated Ags is a function
of DC and not of M. Furthermore, they proposed that by competing for
Ag uptake, M actually inhibit DC cross-presentation. It should be
noted that this model contrasted with the ability of M to present on
MHC class I and also on MHC class II molecules other types of exogenous
Ags (not cell-associated) such as soluble proteins, proteins bound to
beads, or recombinant proteins expressed by bacteria
(24, 25, 26, 27) and to reconstitute CTL responses in vivo
(16). However, the view that M cannot cross-present Ags
supplied by virus-infected cells is very widely held.
To gain a better understanding of the mechanisms of cross-presentation
during the course of a viral infection, we developed an in vitro assay
of MHC class I Ag cross-presentation where Ag production is restricted
to vaccinia virus-infected ADC, and Ag presentation is restricted to
pAPC (see Fig. 1
for a diagram of the
assay). Using this assay, we show that M as well as DC are capable
of cross-presenting viral Ags supplied by infected cells. This
cross-presentation is the direct result of viral infection as it occurs
in the absence of further treatment of the ADC to induce apoptosis,
necrosis, or to inactivate the virus. In addition, we show that our
model Ag follows the cytosolic route within the pAPC, since it requires
proteasome processing, TAP transport, and is inhibited by brefeldin A.
Moreover, we also found that M and DC cross-present Ag with fast
kinetics since cross-presentation is detectable within 1 h of
ADC-pAPC encounter, peaks between 2 and 5 h, and decreases rapidly
to almost nil within the next 10–15 h. Strikingly, cross-presentation
does not require apoptosis or necrosis of the ADC, since live-infected
cells are capable of supplying Ags to pAPC within few hours of becoming
infected.
|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All cells were maintained in complete RPMI media (CRPMI) that
consisted of RPMI supplemented with 10% FCS (Atlanta Biologicals,
Norcross, VA), 2 mM L-glutamine, penicillin-streptomycin,
0.01 M HEPES buffer and nonessential amino acids (all from Invitrogen,
Carlsbad, CA), and 5 x 10-5 2-ME
(Sigma-Aldrich, St. Louis, MO). A9 cells (H-2K)
are an APRT and HPRT negative derivative of strain L cells (no.
CCL-1.4; American Type Culture Collection (ATCC), Manassas, VA). A9-T7
cells (28, 29) are A9 cells stably transfected with T7
polymerase (a kind gift from Dr. B. Moss, National Institutes of
Health, Bethesda, MD). B3Z (30) is a CTL hybridoma that
produces 
-galatosidase (-gal) upon recognition of the OVA
epitope SIINFEKL in the context of the H-2Kb
molecule (a kind gift from Dr. N. Shastri, University of California,
Berkeley, CA). All cells were grown and assays incubated at 37°C in
an atmosphere of 5% CO2 unless otherwise
indicated.
Mice
All experiments using animals were performed under protocols approved by the Institutional Animal Use and Care Committee. All the mice used in the experiments were bred at Fox Chase Cancer Center’s Laboratory Animal Facility (Philadelphia, PA). TAP0/0 (B6-Tap1tp1Arp) breeders were originally purchased from The Jackson Laboratory (Bar Harbor, ME). C57BL/6 (B6) mice were from the Fox Chase Cancer Center colony.
Viruses
The WR strain of vaccinia virus and WR expressing the phage T7 polymerase was a gift from Dr. B. Moss. Vaccinia virus expressing the minigene (M)SIINFEKL was a kind gift of Dr. J. Yewdell (National Institutes of Health). All vaccinia stocks were grown in Hela S3 cells (no. CCL-2.2; ATCC) in T150 tissue culture flasks and virus titers determined using BS-C-1 cells (no. CCL-26; ATCC) growing in 6-well plates following published procedures (29). Virus titers were adjusted to 109 PFU in CRPMI.
Plasmids
We generated a PCR fragment of OVA197–386 preceded by a Kozac sequence and ending in a stop codon using as template a plasmid containing the full cDNA of OVA (a generous gift from Dr. L. Shen, University of Massachusetts Medical Center, Worcester, MA). As forward and reverse primers we used oligos with the sequences CACCATGCCTTTCAGAGTGACTGAGCA and TTAAGGGGAAACACATCTGCC, respectively. The PCR fragment was directionally cloned into pCDNA 3.1-TOPO (Invitrogen) following manufacturer’s instructions to obtain pCDNA-OVA197–386. The cloned fragment was excised from pCDNA-OVA197–386 with BamHI and NotI and inserted into the BamHI-NotI sites of plasmid pBlueScript II SK (pBS; Stratagene, La Jolla, CA) to generate pBS-OVA197–386. Transcription and translation from OVA197–386 generates a fragment of chicken OVA (residues 197–386) lacking the endoplasmic reticulum (ER) transfer signal sequence but containing the Kb-restricted OVA epitope SIINFEKL (amino acid single letter code) corresponding to positions 258–265.
ELISA
The presence of OVA197–386 was determined by a sandwich ELISA in lysates of ADC that had been prepared as described in the next section and lysed at 106 cells/ml in PBS 0.1% Triton X-100 buffer. For the ELISA, 96-well flat-bottom RIA/ELISA plates (Costar; all other plasticware was from BD Biosciences, Mountain View, CA) were incubated overnight with 50 µl of a 1/2,500 solution of goat anti-OVA Ab (ICN Pharmaceuticals, Costa Mesa, CA), washed, and blocked with 200 µl ELISA buffer (PBS 0.5% nonfat milk, 0.1% Triton X-100, 0.025% Tween 20) for 2 h at 37°C. A total of 50 µl of cell lysates in ELISA buffer were added to triplicate wells as sets of seven 10-fold serial dilutions and incubated at room temperature for 3 h. The plates were thoroughly washed and 50 µl of a 1/16,000 dilution of rabbit anti-OVA (ICN Pharmaceuticals) diluted in ELISA buffer was added and incubated at room temperature for 2 h. The plates were washed again, incubated with 1/8,000 dilution of goat anti-rabbit IgG conjugated with HRP (Kirkegaard & Perry Laboratories, Gaithersburg, MD) in ELISA buffer, washed, incubated with tetramethylbenzidine substrate (Kirkegaard & Perry Laboratories) for 20 min and read at 450 nm in a µ Quant 96-well spectrophotometer (Bio-tek Instruments, Watford Herts, U.K.). As a positive control, we used purified chicken OVA (ICN Pharmaceuticals). However, the exact quantification of the amount of OVA197–386 present in the cell lysates was not possible because the polyclonal Abs that we used were raised against full-length OVA. Therefore, some epitopes present in the full-length OVA might be absent in OVA197–386, which may skew the signal. As a guide, the absorbance for 10 pg/well of purified OVA was 0.259 at 450 nm of Abs. All ELISA experiments were repeated at least three times. Each data point represents the mean of triplicate wells for one of the dilutions. The value of the signal decreased linearly with the decrease in the concentration of the cell lysate or of the standard.
Ag cross-presentation assay
Preparation of ADC A9-T7 or A9 cells were seeded in 60-mm tissue culture dishes (1.2 x 106 cells/dish) or 6-well plates (4 x 105 cells/well). Following overnight incubation at 37°C, the cells were transfected with the indicated DNA using Lipofectamine 2000 (Invitrogen) as recommended by the manufacturer. Four or 6 h later, the monolayer was washed twice with PBS and the cells infected for 2 h with 10 PFU/cell (the number of cells was an estimation calculated as twice the initial number of cells) of the indicated vaccinia virus in Optimem media (Invitrogen). Next, the monolayer was washed twice with complete media and incubated in CRPMI for the indicated times. When indicated, the cells were irradiated for 5 min with a short wave (254 nm) Spectroline germicidal UV lamp EF180 (Spectronics, Westbury, NY) placed at 15 cm over the cells. The cells were harvested with a rubber policeman, centrifuged, and the pellet resuspended to an estimated concentration of 106 or 107 cells/ml according to the requirements of the particular experiment.
In some experiments (see Fig. 4), the ADC cells were split in aliquots
and frozen at -80°C. Aliquots were thawed at different times,
vortexed, and used in the cross-presentation assay. In other
experiments (see Fig. 5), the cells were transfected with the indicated
PBS-OVA197–386 and the infections were performed
at different times, the first infection being 4 h before
transfection and the last infection just immediately before harvesting.
In this way, all the ADC cultures containing cells that had been
transfected at the same time but infected at different times were
harvested together.
|
|
To obtain DC, bone marrow cells obtained from the C57BL/6 or
TAP0/0 mice were incubated overnight in CRPMI.
Nonadherent cells were collected, seeded at 106
cells/ml, and cultured for 5–6 days in CRPMI supplemented with 7 ng/ml
GM-CSF (BD PharMingen, San Diego, CA) which was replaced every 2–3
days (BD PharMingen). To obtain M, the procedure was similar, but
instead of GM-CSF, the media contained 20% of L cell supernatant as a
source of M-CSF (16).
Cross-presentation assays
In some experiments, pAPC were grown in 10-mm bacteriological
grade petri dishes (BD Biosciences), collected with a rubber policeman,
washed, counted, and 105 placed at the indicated
ADC:pAPC ratio in duplicate wells of white, flat-bottom 96-well
plates (BD Biosciences). In other experiments, pAPC were grown
in 24-well plates and cross-presentation was initiated by adding 5
x 105 ADC (107 cells/ml).
In some cases, 20 min before adding the ADC, the media of the pAPC was
replaced with 200 µl of CRPMI that contained 16 µM lactacystin, 1
µg/ml brefeldin A (both from Sigma-Aldrich), or diluent. In this case
ADC were added directly, without media replacement. To measure direct
presentation by M and DC infected with vaccinia (M)SIINFEKL, 5
PFU/cell of virus stock was added instead of the ADC. To allow for
cross-presentation or direct presentation, the ADC-pAPC cocultures or
infected pAPC were incubated for 4 h or the indicated times,
harvested using the plunger of a 1 ml syringe as a rubber policeman.
When indicated, the harvested cells were spun and fixed for 15 min in
200 µl of 0.5% paraformaldehyde, washed twice with PBS, neutralized
for 20 min in 0.5 mM L-lysine in CRPMI, washed twice in
CRPMI, and serially diluted in white 96-well plates. In some cases, the
procedure was similar but the cells were not fixed. The concentration
of pAPC in these conditions was estimated by counting the number of
pAPC in wells that did not receive ADC. In other experiments, the
process was reversed and the pAPC were added to the ADC cultures. For
this purpose, the ADC were prepared by transfecting and infecting A9-T7
cells in 6-well plates. After 9 h, any detached ADC cells were
removed by gently washing the monolayer with CRPMI after which
106 pAPC that had been grown in 100-mm petri
dishes were added to the ADC that remained attached to the wells.
Following a 4-h incubation, the pAPC-ADC cocultures were harvested
using a rubber policeman, fixed, and seeded in 96-well plates as
before.
To determine cross-presentation, 105 B3Z
responder cells were added to each well of the 96-well plates above and
incubated for an additional 16–24 h to allow for hybridoma activation
and -gal production. To detect -gal activity, we used the
Galactostar chemiluminescent kit (Applied Biosystems, Foster City, CA)
following the instructions of the manufacturer except that we used 15
µl/well of lysis buffer and 40 µl/well of substrate diluted 1/150
in reaction buffer. The assays were read using a Topcount instrument
(Packard Instrument, Meriden, CT).
All experiments of cross-presentation were repeated at least three
times and each figure corresponds to a representative experiment. Each
data point in the figures represents the average of duplicate wells.
Positive controls included pAPC incubated with 2-fold serial dilutions
of SIINFEKL at an initial concentration of 200 or 100 pM and a final
concentration of 3.2 or 1.56 pM. This range of concentrations induced a
nonsaturating, linear response by the hybridoma. The maximal
stimulation through cross-presentation fluctuated around the values
obtained between 3.12 and 6.25 pM and never reached the -gal
activity induced with 50 pM SIINFEKL.
Ab staining for flow cytometry
For cell surface staining 0.5 x 106 cells were incubated in a well of a U-bottom 96-well microtiter plate with 0.1 µg of the corresponding primary labeled Ab in 50 µl FSB (PBS 2% FCS, 0.04% sodium azide). The following primary labeled Abs used were PE-labeled anti-MHC class II (I-Ab-specific clone AF6-120.1), FITC-labeled anti-CD11c (clone HL3), FITC-labeled anti-Vb8.1,8.2 (clone MR5-2), PE-labeled anti-CD8 (clone 53-6.7) from BD PharMingen, and FITC-labeled anti-CD11b (a kind gift of Dr. R. Hardy, Fox Chase Cancer Center). For immediate analysis, the cells were washed with FSB three times and resuspended in 0.4 ml SB. When the analysis was postponed, the cells were washed three times with PBS and resuspended in 0.2 ml PBS containing 0.5% paraformaldehyde.
Detection of cell death
Ethidium bromide and acridine orange staining for microscopy ADC well prepared in 6-well plates as above. After a 13-h infection, 10 µl of ethidium bromide (Sigma-Aldrich) and acridine orange (Fisher Scientific, Pittsburgh, PA) (both stocks at 100 µg/ml in PBS) were added, and the cells were immediately observed under FITC (green) and Rhodamine (red) filters using a Nikon Diaphoto microscope (Nikon, Melville, NY). Pictures were taken with a Pixera Pro ISOES camera (Nikon) coupled to the Nikon microscope at x100 magnification. The green and red images were merged using Picture Publisher 10.0 software (Micrograph, Dallas, TX).
TUNEL assay for apoptotic death
Apoptotic death was assessed using the TUNEL kit (Roche Diagnostic Systems, Somerville, NJ). Briefly, ADC were prepared in 6-well plates, and following a 13-h infection ADC were washed in PBS, fixed in 4% paraformaldehyde, and permeabilized in 0.1% Triton X-100, 0.1% sodium citrate for 2 min on ice. Fixed and permeabilized ADC were washed with PBS and incubated for 1 h at 37°C in a solution containing 25 mM Tris (pH 6.6), 200 mM cacodylate, 1 mM CoCl2, 0.6 nM fluorescein-12dUTP, and 25 U of TdT (Roche Diagnostic Systems). ADC were analyzed by fluorescent microscopy in a drop of PBS using a FITC band path filter.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and DC can cross-present Ags shed by vaccinia-infected
cells
To study cross-presentation in viral infections, it is absolutely
necessary to restrict Ag synthesis to the ADC and Ag presentation to
the pAPC. Vaccinia can infect pAPC and cannot be used to study
cross-presentation as a consequence of viral infection if the Ag is
encoded within the viral genome because direct presentation would mask
cross-presentation. Therefore, to study cross-presentation as a
consequence of vaccinia virus infection, we restricted Ag synthesis to
the ADC by adapting a method originally used to produce recombinant
proteins in mouse cells. In this system, A9-T7 cells, a derivative of
the mouse fibroblast cell line A9 (an HPRT and APRT negative mouse
subclone of the L strain, H-2K) stably expresses
the bacteriophage T7 polymerase. These cells produce a protein encoded
within a transiently transfected plasmid and controlled by the T7
promoter only under conditions of vaccinia virus infection
(28). The ability of vaccinia virus to induce expression
of a T7 controlled protein in the presence of T7 polymerase has been
attributed to the capacity of the virus to stabilize the plasmid
encoded RNA in the cytoplasm. For our purposes, we transfected A9-T7
cells with pBS-OVA197–386 and infected with
wild-type vaccinia virus. The cells were then incubated for 24 h,
harvested, lysed, and the presence of OVA197–386
was determined by ELISA. Fig. 2A shows that A9-T7 cells that
had been transfected with pBS-OVA197–386 and
infected with wild-type vaccinia produced large amounts of
OVA197–386 while transfection of A9-T7 cells
with pBS-OVA197–386 in the absence of vaccinia
infection or parent A9 cells (T7 negative) that had been transfected
with PBS-OVA197–386 and infected with wild-type
vaccinia did not produce OVA197–386. Further
controls showed that A9 cells produced as much OVA as A9-T7 cells when
they were transfected with pBS-OVA197–386 and
infected with recombinant vaccinia virus carrying the T7 polymerase,
indicating that parent A9 cells were an appropriate negative control
since they could be transfected and infected with similar efficiency as
A9-T7 cells. In addition, OVA197–386 was not
detected in lysates of A9-T7 cells that had been infected with vaccinia
wild type, but transfected with vector alone (data not shown). In
related experiments, we transfected A9-T7 and A9 cells with pBS--gal
instead of pBS-OVA197–386 to find that most
(80–100%, varying with each experiment) of the A9-T7 cells that had
been infected with vaccinia wild type stained blue very rapidly (within
a few minutes) and very strongly with X-gal while no staining was found
in A9 cells infected with wild-type vaccinia (Fig. 2B). In
addition, strong staining was observed in control A9-T7 and A9 cells
that had been infected with vaccinia-T7 but not in uninfected
A9-T7 cells (data not shown). Therefore, A9-T7 cells transfected with
pBS-OVA197–386 and infected with wild-type
vaccinia are an excellent source of ADC to study cross-presentation. In
this system, OVA197–386 constitutes a model of a
viral Ag as it is only produced upon vaccinia infection. In addition,
ADC irradiation is not required because the transfer of live virus to
the pAPC or even unwilling transfection of the pAPC by carryover
plasmid (an unlikely but theoretically possible problem that may occur
when using vaccinia-T7 instead of A9-T7 cells) should not result in
production and direct presentation of OVA197–386
by the pAPC.
|
and DC
prepared from bone marrow cells obtained from C57BL/6 (B6) mice. The
identity of the pAPC was confirmed by microscopic morphology (data not
shown) and by flow cytometry, which indicated that the bone marrow
cells cultured in the presence of M-CSF expressed MHC class II but not
CD11c as is expected for M, while the cells cultured in the presence
of GM-CSF expressed MHC class II and CD11c as expected for DC (Fig. 2
C). It should be noted that we consistently found that the
M cultures were highly homogeneous, while the DC were more
heterogeneous with day to day variations (40–70%) in the proportion
of cells expressing CD11c. In addition, both cell types expressed
CD11b, CD16, H-2Kb, and the M scavenger
receptor A. DC expressed moderate levels of B7-2 and low B7-1 while
M expressed low B7-2 and no B7-1. Both cell types increased MHC and
B7 expression upon activation with LPS or UV- irradiated
vaccinia-infected A9-T7 cells (data not shown). To determine
cross-presentation, A9-T7 or A9 control ADC
(H-2K) were transfected and infected as indicated
in Materials and Methods, harvested, washed, and plated at
different concentrations in 96-well plates. To these cells, B6-derived
(H-2b) pAPC (M or DC) and B3Z responder T
cells (OVA-specific, H-2Kb-restricted) were
added. The cells were incubated overnight and the activation of B3Z was
determined by measuring -gal activity. Our results show that both
M and DC cross-presented OVA197–386 when
cocultured with A9-T7 cells that had been transfected with
pBS-OVA197–386 and infected with wild-type virus
(Fig. 2D). Importantly, no signal was detected when we used
similarly treated A9 cells that did not produce
OVA197–386 as ADC controls. It should be noted
that in this experiment as well as others, the cross-presentation
signal obtained with M was higher than with DC, typical of most but
not all our experiments. Although an interpretation could be that M
cross-present better, it is more likely that the day to day variation
in the purity of the DC cultures results in this difference. In
additional controls, A9 cells and A9-T7 cells transfected with
pBS-OVA197–386 and infected with recombinant
vaccinia-T7 cross-presented similarly (data not shown). Also, no signal
was detected when A9-T7 cells were infected with wild-type vaccinia and
transfected with empty pBS vector (see Fig. 2E).
Furthermore, when we used ADC supernatants no cross-presentation was
observed indicating that the cross-presented Ag is cell associated (see
Fig. 2E). Altogether, these experiments demonstrate that
M can cross-present viral Ags from vaccinia-infected cells similar
to DC, and that cross-presented Ags are cell associated. It also
demonstrates that vaccinia-infected cells provide Ag for
cross-presentation without a requirement to artificially induce
apoptosis or necrosis of the infected cells. This cross-presentation is
highly efficient as only 0.06 ADC are required for each pAPC to detect
cross-presentation, and 0.25 ADC per pAPC provides saturating amounts
of Ag (Fig. 2D). Cross-presentation of OVA197–386 follows the cytosolic route
Previous work with proteins bound to beads and bacteria expressing
recombinant proteins led respectively to the discovery of the cytosolic
and vacuolar routes of exogenous Ag presentation (26, 31).
The cytosolic route is characterized by Ags gaining access to the
cytosol of the pAPC and, similar to the presentation of endogenous
proteins (32, 33, 34, 35, 36), the Ags are degraded by the proteasome
and the resulting peptides transported by the TAP to the ER. In the ER,
the peptides bind to newly synthesized MHC class I molecules and are
transported to the cell surface via the secretory pathway. In the
vacuolar route, the exogenous Ags do not gain access to the cytosol and
are most likely degraded by proteases within endosomal or lysosomal
compartments and the resulting peptides bind to MHC class I molecules
that recycle from the cell surface (26, 37). We have
previously shown that bone marrow chimeric mice with TAP-deficient pAPC
were inefficient in generating CTL responses to several viruses,
including vaccinia, suggesting that the dominant route of
cross-presentation in viral infections is cytosolic (2, 3). However, the use of TAP deficiency as the single method to
demonstrate that an Ag follows the cytosolic route has been criticized
because TAP deficiency reduces the number of MHC class I molecules at
the cell surface that could be available for recycling and could
indirectly affect the vacuolar route (38). Therefore,
these results may require confirmation in a more controlled environment
where each of the steps of the cytosolic route can be individually
evaluated. We first tested the role of TAP in the cross-presentation of
Ags supplied by vaccinia-infected cells. For this purpose, A9-T7 ADC
were prepared as before and pAPC were obtained from either
TAP-deficient or wild-type B6 mice as controls. As seen in Fig. 3A, M and DC from wild-type
B6 mice were able to cross-present OVA197–386
delivered by vaccinia-infected cells while those from TAP-deficient
mice did not cross-present, strongly suggesting a dominant role for the
cytosolic route. To confirm this, we blocked other steps of the
cytosolic route by treating B6 M and DC with lactacystin to inhibit
the proteasome or with brefeldin A to block the secretory pathway. To
minimize any toxic or nonspecific effects of the drugs on the pAPC or
the T cell hybridoma, the pAPC were only exposed to the inhibitors for
20 min before the incubation with the ADC and during the 4 h of
the processing period. After this, the pAPC were fixed with
paraformaldehyde and thoroughly washed before mixing with B3Z cells.
These experiments showed that both lactacystin and brefeldin A
completely inhibited cross-presentation (Fig. 3B). As a
control, we infected pAPC with recombinant vaccinia virus expressing
the minigene (M)SIINFEKL which does not require proteasome processing
but requires de novo MHC class I synthesis. As expected, we found that
brefeldin A but not lactacystin inhibited direct presentation by M
or DC (Fig. 3B, small insets). In additional
controls we found that M and DC still cross-presented when they were
not fixed and brefeldin A was removed by thorough washing before adding
the B3Z responder cells (Fig. 3B). Altogether, these data
demonstrate that M and DC use the cytosolic route to cross-present
OVA197–386 delivered by vaccinia-infected
cells.
|
and DC is rapid and transient
To address how soon and for how long pAPC cross-present Ag
following pAPC-ADC contact, we prepared ADC as before although they
were frozen in aliquots which, at different times points within a 16-h
period, were thawed, and saturating numbers (1 ADC:1 pAPC) were
immediately added to cultures of M or DC. All ADC-pAPC cocultures
were harvested together at the end of the assay, fixed with
paraformaldehyde to stop further Ag processing, thoroughly washed, and
serially diluted in duplicate wells of 96-well plates to which B3Z
responder cells were added. Cross-presentation was determined following
overnight incubation. The results indicated that M and DC
cross-presented Ag acquired from vaccinia-infected cells with fast and
similar kinetics (Fig. 4A).
Maximal cross-presentation by both pAPC was reached between 2 and
5 h of ADC-pAPC contact and decreased thereafter with very little
Ag remaining at the surface of pAPC after 16 h of ADC-pAPC
coculture. It was reasonable to think that cross-presentation
extinguished relatively quickly because vaccinia virus carried over by
the ADC may have infected the pAPC and may have affected their ability
to present Ags at later time points adversely. To test for this
possibility, we performed a set of experiments where carry-over virus
was inactivated by irradiating the ADC with UV light before dividing in
aliquots and freezing while all other conditions remained the same.
This irradiation rendered ADC that did not transfer virus since their
lysates did not infect BSC-1 cells nor did they induce expression of
-gal in A9-T7 that had been transfected with pBS--gal while
lysates of infected, nonirradiated controls did (data not shown). In
addition, M and DC that had been incubated with irradiated ADC for
24 h remained viable and became activated as revealed by increased
expression of MHC and B7 molecules by FACS similar to the activation
induced by incubation with LPS (data not shown). Despite virus
inactivation, the kinetics of cross-presentation remained the same
(Fig. 4B). From these results we conclude that M and DC
cross-present viral Ag acquired from vaccinia-infected cells rapidly
and transiently in the presence or in the absence of active virus.
However, it should be noted that in these last experiments the ADC were
obviously dead as a consequence of the freeze-thaw and, in Fig. 4B, the irradiation. Nonetheless, this was the only way to
keep our Ag source identical from one data point to the other without
having to process each set of cells at different times and storing the
fixed pAPC for prolonged times.
ADC provide Ag for cross-presentation at early times of infection
Another important aspect in the kinetics of cross-presentation was
to determine how soon after infection vaccinia-infected ADC transfer Ag
to pAPC. For this purpose, we transfected A9-T7 cells with
pBS-OVA197–386 as in previous experiments, but
the cells were infected with wild-type vaccinia at different times. All
ADC that had been infected for different times were collected at the
same time and added to cultures of M or DC. After a 4-h incubation
at 37°C, the ADC-pAPC cocultures were collected, fixed with
paraformaldehyde to stop any additional cross-presentation, thoroughly
washed, and serially diluted into duplicate wells of 96-well plates to
which B3Z responder cells were added. Our results show that
vaccinia-infected ADC could transfer Ag to M after only 5 h of
infection, whereas DC were somewhat slower (Fig. 5, unbroken lines and left
ordinate). Maximal cross-presentation was reached when the ADC had
been infected for 9 h. Interestingly, the kinetics of
cross-presentation followed the kinetics of Ag expression by the ADC as
determined by ELISA (Fig. 5, dotted line and right
ordinate). Altogether, these data indicate that ADC provide Ag for
cross-presentation at early times of infection. That vaccinia-infected
ADC can deliver Ag to pAPC at this very early time of the infectious
process is surprising because at this stage ADC death should be
minimal.
Cross-presentation does not require death of the ADC
Death of the ADC by either apoptosis or necrosis is one of the
hallmarks of the current model of cross-presentation and has been
argued to play a principal role in the recognition of ADC by pAPC. That
pAPC can cross-present Ags from dead cells is confirmed by our
experiments in
Figs. 2–4 where many or all of the ADC were dead
because they were used at late stages of infection and in Fig. 4, they
were freezed and thawed and even irradiated. However, while vaccinia is
a highly cytopathic virus that exerts very early morphological changes
in infected cells (39), it has also developed strong
antiapoptotic strategies (40, 41, 42) and death of
vaccinia-infected cells should not occur for many hours after
infection. The experiments in Fig. 5 showing cross-presentation at
early times after infection suggested that cell death might not be
required for the transfer of Ag from vaccinia-infected ADC to pAPC.
However, the conditions for this experiment included harvesting the
ADC, which could have induced some degree of ADC death. To carefully
test whether live-infected cells could serve as a source of Ag for
cross-presentation, we modified our assay to avoid harvesting the ADC
before cross-presentation. For this purpose, we prepared ADC by
transfecting and infecting A9-T7 in 6-well plates. Nine hours after
infection, the ADC monolayer was gently washed with CRPMI to remove all
detached ADC. Next, pAPC (to determine cross-presentation) or CRPMI (to
determine cell viability) were added to the ADC that remained attached
to the wells. After a 4-h incubation, the ADC-pAPC cocultures were
harvested, immediately fixed, and cross-presentation determined as
usual. Fig. 6A shows that both
M and DC cells effectively cross-presented Ag under these
conditions. At the same time that the ADC-pAPC cocultures were
harvested, the cells that received only CRPMI were stained with
acridine orange and ethidium bromide or using the TUNEL assay. Acridine
orange is a supravital dye that intercalates in the DNA and stains the
nuclei of all cells green. With this staining, apoptosis can be
determined by comparing the morphology of the stained nuclei. Ethidium
bromide stains the nuclei of nonviable cells bright orange. Fig. 6B demonstrates that A9-T7 cells that had been transfected
and infected 13 h earlier (this time includes the 9 h of
infection before adding the ADC to the cross-presentation wells and the
4 h of pAPC/ADC coculture) exhibited the typical rounding and
membrane roughness of vaccinia-infected cells. However, they were
>97% viable as revealed by exclusion of ethidium bromide, and they
were not apoptotic as revealed by the morphology of the nuclei and the
absence of nuclear fluorescence in the TUNEL assay. For comparison, the
staining of normal, apoptotic, and saponin-permeabilized apoptotic
cells are included in the figure. The low level of death (<3%) cannot
explain the cross-presentation observed, because these ADC induced
cross-presentation signals similar to those where 100% of the cells
were dead. Therefore, live-infected ADC can transfer Ag to pAPC for
cross-presentation.
|
indicated that dead cells could transfer Ag to pAPC
following only 1 h of coculture, we tested whether the same
occurred when using live-infected cells as ADC. Therefore, A9-T7 cells
were prepared as in Fig. 6 and cocultured with pAPC for 1 or 2 h.
The results in Fig. 7 show that, similar
to what occurs with dead-infected ADC, pAPC can cross-present Ag from
live-infected ADC within 1 h of coculture.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
can also cross-present Ag supplied by vaccinia-infected ADC with
high efficiency. Moreover, our results show that for the mechanisms and
kinetics of cross-presentation that we have analyzed, M and DC are
almost identical. In the past few years, there has been much emphasis
placed on the exclusive role of DC as the primary cell type involved in
Ag presentation while the idea of M as major player in this process
has been almost abandoned. One possible reason for our results being
different from those reported by Albert and coworkers (21, 22) is that human M and mouse M may differ in their
ability to cross-present. It is also possible that M of differing
tissue origin (peripheral blood monocyte-derived vs bone
marrow-derived) differ in their ability to cross-present. Another
possibility for this difference is that different pAPC could handle
diverse Ags or cells infected by various viruses differently since we
used OVA and vaccinia virus while Albert et al. (21, 22)
used influenza virus. This notion is reinforced by the demonstration
that M generated under similar conditions as ours failed to present
exogenous OVA in the form of immune complexes (44).
Nonetheless, our finding that M cross-present Ags contradicts a
model generalizing the view that cross-presentation of Ags provided by
infected cells is an exclusive ability of DC and that M are
scavengers with an enzymatic machinery that destroys these Ags rather
than uses them to generate antigenic peptides for MHC class I
cross-presentation (22). Therefore, the role of M as
pAPC should be revisited. However, it should be noted that our assay
measures Ag presentation and not priming of naive
CD8+ T cells. Therefore, it remains possible that
the activation of naive cells in the lymph nodes is an exclusive
function of DC. This would agree with the observation that DC but not
M can be found in the T cell areas of lymphoid organs
(45). However, cross-presenting M could be involved in
the amplification of the response in the periphery, or in the rapid
activation of memory cells in tissues during recall responses.
Alternatively, there is the provocative possibility that
cross-presenting M that localize to the subcapsular region of lymph
nodes (17, 45) could rapidly differentiate to DC and
migrate to the lymph nodes to activate naive T cells. In this manuscript, we also show that OVA197–386 as a model of a vaccinia Ag is cross-presented by pAPC using the cytosolic route. This result agrees and confirms our previous work in vivo using poliovirus also expressing a truncated, cytosolic form of OVA, and with the results from other laboratories using a different virus and Ag (22). Nevertheless, it is also interesting to note that for Ags delivered in the form of bacteria or protein bound to microbeads, exogenous presentation of SIINFEKL followed the cytosolic route when expressed in the context of the native OVA sequence (as in our experiments). However, it could use the vacuolar pathway when expressed in a recombinant form, fused very close to the C terminus of the Escherichia coli Crl protein. Whether these findings are mirrored when the exogenous Ag is provided by vaccinia infected cells is currently under investigation (26).
The current model of cross-presentation during viral infections is that DC acquire Ag from apoptotic- or necrotic-infected cells in the periphery and migrate to lymphoid organs (13, 46, 47). This process is accompanied by the maturation of the DC, which involves a decreased ability to phagocytose and increased expression of MHC, B7, and other molecules allowing for a more effective presentation of Ag to T cells (48, 49). Our experiments show that pAPC cross-present Ags released by vaccinia-infected cells very soon after contacting ADC, that cross-presentation lasts for a short period of time, and that cross-presentation starts at early times of infection when the ADC are still alive.
That pAPC cross-present Ag from virus-infected cells very soon after
ADC-pAPC contact is in agreement with the finding that cross-presenting
cells migrate to the draining lymph nodes within 6 h after
scarification of the epidermis with HSV (50).
However, our finding that cross-presentation of Ags from
vaccinia-infected cells is also short-lived even in the presence of
saturating amounts of Ag and in the absence of pAPC infection is
surprising. Indeed, it would be expected that direct presentation by
pAPC infected with cytopathic viruses would be short-lived but not
during cross-presentation when the pAPC infection is spared as when we
irradiated the ADC. Still, it is very likely that our results reflect
the duration of cross-presentation for individual cells in vivo, and is
consistent with the measured half-life of MHC class I peptide complexes
of high affinity at the surface of cells, which is 
6–12 h
(51, 52). If this is correct, it implies that the
activation of pAPC and their ability to stimulate naive T cells should
also occur very fast, and that the activation of a sufficient number of
precursor CTL in vivo may require the continuous arrival of pAPC from
the site of infection. An alternative explanation for the short time of
cross-presentation during vaccinia infection may be that vaccinia may
produce cross-presentation inhibitory factors. Poxviruses are known to
encode multiple immune evasion molecules such as cytokine and
chemokine-like proteins (53, 54). If this were the case,
the immune evasion molecules acting in our assay should be transferred
with the infected cells because we discarded the supernatants.
Moreover, these factors should work in trans (i.e., produced
in the ADC but functions in the pAPC) because virus inactivation did
not alter the kinetics of cross-presentation. To our knowledge,
inhibition of MHC class I Ag presentation in trans has not
yet been described. If this were the reason for the short time that
cross-presentation lasts, it might be of interest to identify the
responsible molecules. Another alternative is that some mechanisms that
prolong Ag cross-presentation in vivo may be absent in our assay. For
example, cytokines or inflammatory molecules could increase the
half-life of the MHC class I peptide complexes at the cell surface
(48) or retard the intracellular degradation of Ag within
the pAPC.
It is widely held that cross-presentation in viral infections required
either necrotic or apoptotic cells as the source of Ag. Therefore, our
present findings that cross-presentation occur at early times of
vaccinia infection and without the need for cell death is striking. It
will now be important to identify the signals used by the infected ADC
to alert pAPC and to determine the mechanisms whereby pAPC incorporate
the Ags supplied by live-infected ADC. One possible mechanism is that
pAPC phagocytose and process whole live-infected ADC. It is also
possible that M and DC cells recognize signals imparted by the
infected cells and kill them just before phagocytosis. Other
possibilities are that live-infected ADC release Ag-loaded exosomes
(55) or antigenic peptides bound to heat shock proteins
(56). These could be taken up by phagocytosis or through
receptor-mediated endocytosis and cross-presented by pAPC. We also
speculate that the ability of pAPC to acquire Ag from live ADC at early
times of infection could be advantageous because at this time the
replication cycle of the virus might not have been completed and pAPC
could avoid becoming infected themselves.
Together, our findings that cross-presentation during vaccinia infection occurs very rapidly following ADC-pAPC contact and that ADC provide Ag before dying and at early phases of infection should contribute to rapid Ag presentation and a headstart in the initiation of the CD8+ T cell response. This timing could be critical to tip the balance in favor of the host in its race against the virus. Together, our findings and those of others (17, 50) point to a model where direct presentation and cross-presentation are both fast and brief. This would also require pAPC-T cell contact to be brief and occur at a very early phase of the immune response. This model would also be in agreement with findings that a short exposure of CD8+ T cells to Ag bearing pAPC is sufficient to start a program of maximal proliferation and differentiation into effector and memory cells (57, 58, 59, 60).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Luis J. Sigal, Associate Member, Fox Chase Cancer Center, 7701 Burholme Avenue, Philadelphia PA 19111. E-mail address: lj_sigal{at}fccc.edu
3 Abbreviations used in this paper: pAPC, professional APC; ADC, Ag donor cell; CRPMI, complete RPMI media; DC, dendritic cell; M, macrophage; -gal, -galatosidase; ER, endoplasmic reticulum.
Received for publication August 20, 2002. Accepted for publication October 11, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
v5 and CD36, and cross-present antigens to cytotoxic T lymphocytes. J. Exp. Med. 188:1359.2-microglobulin in the processing of bacterial or particulate antigens via an alternate class I MHC processing pathway. J. Immunol. 156:4182.[Abstract]
-converting enzyme and protects virus-infected cells from TNF- and Fas-mediated apoptosis, but does not prevent IL-1-induced fever. J. Gen. Virol. 78:677.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  X. Ma, A. Serna, R.-H. Xu, and L. J. Sigal The Amino Acid Sequences Flanking an Antigenic Determinant Can Strongly Affect MHC Class I Cross-Presentation without Altering Direct Presentation J. Immunol., April 15, 2009; 182(8): 4601 - 4607. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. Jirmo, C.-H. Nagel, C. Bohnen, B. Sodeik, and G. M. N. Behrens Contribution of Direct and Cross-Presentation to CTL Immunity against Herpes Simplex Virus 1 J. Immunol., January 1, 2009; 182(1): 283 - 292. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. R. Mott and H. Ghiasi Role of Dendritic Cells in Enhancement of Herpes Simplex Virus Type 1 Latency and Reactivation in Vaccinated Mice Clin. Vaccine Immunol., December 1, 2008; 15(12): 1859 - 1867. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Mouries, G. Moron, G. Schlecht, N. Escriou, G. Dadaglio, and C. Leclerc Plasmacytoid dendritic cells efficiently cross-prime naive T cells in vivo after TLR activation Blood, November 1, 2008; 112(9): 3713 - 3722. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Mott, D. J. Brick, N. van Rooijen, and H. Ghiasi Macrophages Are Important Determinants of Acute Ocular HSV-1 Infection in Immunized Mice Invest. Ophthalmol. Vis. Sci., December 1, 2007; 48(12): 5605 - 5615. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Lundberg, S. K. Drexler, C. Monaco, L. M. Williams, S. M. Sacre, M. Feldmann, and B. M. Foxwell Key differences in TLR3/poly I:C signaling and cytokine induction by human primary cells: a phenomenon absent from murine cell systems Blood, November 1, 2007; 110(9): 3245 - 3252. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Megiovanni, F. Sanchez, M. Robledo-Sarmiento, C. Morel, J. C. Gluckman, and S. Boudaly Polymorphonuclear neutrophils deliver activation signals and antigenic molecules to dendritic cells: a new link between leukocytes upstream of T lymphocytes J. Leukoc. Biol., May 1, 2006; 79(5): 977 - 988. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L.-A. M. Pozzi, J. W. Maciaszek, and K. L. Rock Both Dendritic Cells and Macrophages Can Stimulate Naive CD8 T Cells In Vivo to Proliferate, Develop Effector Function, and Differentiate into Memory Cells J. Immunol., August 15, 2005; 175(4): 2071 - 2081. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Mwangi, W. C. Brown, G. A. Splitter, Y. Zhuang, K. Kegerreis, and G. H. Palmer Enhancement of antigen acquisition by dendritic cells and MHC class II-restricted epitope presentation to CD4+ T cells using VP22 DNA vaccine vectors that promote intercellular spreading following initial transfection J. Leukoc. Biol., August 1, 2005; 78(2): 401 - 411. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Basta, R. Stoessel, M. Basler, M. van den Broek, and M. Groettrup Cross-Presentation of the Long-Lived Lymphocytic Choriomeningitis Virus Nucleoprotein Does Not Require Neosynthesis and Is Enhanced via Heat Shock Proteins J. Immunol., July 15, 2005; 175(2): 796 - 805. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Wahid, M. J. Cannon, and M. Chow Virus-Specific CD4+ and CD8+ Cytotoxic T-Cell Responses and Long-Term T-Cell Memory in Individuals Vaccinated against Polio J. Virol., May 15, 2005; 79(10): 5988 - 5995. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Janda, P. Schoneberger, M. Skoberne, M. Messerle, H. Russmann, and G. Geginat Cross-Presentation of Listeria-Derived CD8 T Cell Epitopes Requires Unstable Bacterial Translation Products J. Immunol., November 1, 2004; 173(9): 5644 - 5651. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Drillien, D. Spehner, and D. Hanau Modified vaccinia virus Ankara induces moderate activation of human dendritic cells J. Gen. Virol., August 1, 2004; 85(8): 2167 - 2175. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Mercier, S. Verhaagh, J. Goudsmit, A. Lemckert, M. Monteil, M. Havenga, and M. Eloit Adenovirus fibre exchange alters cell tropism in vitro but not transgene-specific T CD8+ immune responses in vivo J. Gen. Virol., May 1, 2004; 85(5): 1227 - 1236. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Maranon, J.-F. Desoutter, G. Hoeffel, W. Cohen, D. Hanau, and A. Hosmalin Dendritic cells cross-present HIV antigens from live as well as apoptotic infected CD4+ T lymphocytes PNAS, April 20, 2004; 101(16): 6092 - 6097. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Ramakrishna, J. F. Treml, L. Vitale, J. E. Connolly, T. O'Neill, P. A. Smith, C. L. Jones, L.-Z. He, J. Goldstein, P. K. Wallace, et al. Mannose Receptor Targeting of Tumor Antigen pmel17 to Human Dendritic Cells Directs Anti-Melanoma T Cell Responses via Multiple HLA Molecules J. Immunol., March 1, 2004; 172(5): 2845 - 2852. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. F. Fonteneau, D. G. Kavanagh, M. Lirvall, C. Sanders, T. L. Cover, N. Bhardwaj, and M. Larsson Characterization of the MHC class I cross-presentation pathway for cell-associated antigens by human dendritic cells Blood, December 15, 2003; 102(13): 4448 - 4455. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Serna, M. C. Ramirez, A. Soukhanova, and L. J. Sigal Cutting Edge: Efficient MHC Class I Cross-Presentation during Early Vaccinia Infection Requires the Transfer of Proteasomal Intermediates between Antigen Donor and Presenting Cells J. Immunol., December 1, 2003; 171(11): 5668 - 5672. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
