| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Cell and Molecular Biology, Section for Immunology, Lund University, Lund, Sweden
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
+, CD8-CD4+, and
CD8-CD4- subsets are defined. This study
characterized the localization, number, and function of these subsets
during acute Salmonella typhimurium infection.
Immunohistochemical and flow cytometric analyses of spleens from mice
orally infected with virulent S. typhimurium revealed
that in situ redistribution and alteration in the absolute number and
function of DC occurred in a subset-specific manner during infection.
CD8-CD4+ DC present at B cell follicle
borders in the spleen of naive mice were absent 5 days
post-Salmonella infection, despite no overall change in
the absolute number of CD8-CD4+ splenic DC.
CD8+ and CD8-CD4- DC were
prominently associated with the red pulp, and the frequency of these
cells increased strikingly 5 days post-Salmonella
infection. Significant quantitative increases in both
CD8+ and CD8-CD4- subsets
were associated with the in situ redistribution. Examination of
Salmonella-infected
TAP1-/-/
2-microglobulin-/-
mice, which lack CD8+ T cells, confirmed the
differential subset-specific modulations in the DC populations both in
situ and quantitatively. Ex vivo intracellular cytokine analysis showed
significantly increased frequencies of CD8+ DC producing
TNF- at days 2 and 5 postinfection. In contrast, CD4+ DC
producing TNF- were transiently increased followed by a significant
reduction. No significant increase in IL-12p40 or IL-10 production by
splenic DC was detected during the first 5 days post-S.
typhimurium infection. Together these data reveal differential
modulation of splenic DC subsets with regard to organization, number,
and cytokine production during the course of acute
Salmonella infection. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- (CD11b+) and
CD8+ (CD11b-)
subpopulations have been identified (4, 5, 6, 7, 8). Recently,
splenic CD8- DC have been further divided
into CD4- and CD4+
populations, with the CD4+ DC constituting a
majority of CD8- DC (8, 9, 10).
Thus, subsets among cells classified as DC based on CD11c expression
may now be thought of not only as CD8+ and
CD8- but also as
CD8+,
CD8-CD4+, or
CD8-CD4-
(11).
Whereas functional differences within the CD4+ vs
CD4- populations of
CD8- DC have not yet been reported,
differences between CD8+ and
CD8- DC subsets have been described. Both
CD8+ and CD8- DC
subsets appear capable of producing IL-12, but the requirements to do
so may differ. For example, CD8+ but not
CD8- DC readily produce IL-12 after
stimulation in vitro or in vivo in response to a soluble extract of
Toxoplasma gondii or inflammatory stimuli (8, 12, 13, 14, 15, 16). IL-12 production by CD8- DC
does occur but seems to have more stringent requirements (12, 16). Furthermore, CD8+ rather than
CD8- cells are the predominant DC population
producing IFN-
in an IL-12-dependent manner in vitro
(17). Recent data also suggest that the
CD8+ vs CD8- DC
subsets may differentially direct the response of
CD4+ T cell priming in vivo into effectors
dominated by a Th1 or Th2 cytokine secretion profile, respectively
(11, 14, 18). The mechanism behind this differential
ability to skew an immune response remains to be clarified. It may,
however, be related to differing profiles of cytokine production by the
CD8+ vs CD8- DC
subsets, as DC from IL-12-deficient mice fail to prime Th1 responses
(14).
Additional features that distinguish CD8+ and
CD8- DC subsets is their capacity to capture
Ag and their localization within secondary lymphoid organs. Although
both isolated splenic CD8- and
CD8+ DC acquire particulate and protein Ags in
vitro, the CD8- population did so more
efficiently than CD8+ DC (8, 19).
However, both DC subsets may perform this function with protein Ag in
vivo (20). In the spleen and Peyer’s patches,
CD8- and CD8+ DC
subsets also localize to distinct regions. Whereas
CD8+ DC preferentially localize to the T
cell-rich areas, CD8- DC primarily associate
with the marginal zones in the spleen or the subepithelial dome of
Peyer’s patches (8, 21, 22, 23).
The localization of splenic DC in vivo is also altered by
administration of LPS or T. gondii extract, suggesting that
these stimuli may induce rapid redistribution of DC from marginal zones
to T cell regions of the spleen (13, 20, 24). LPS also
induced DC maturation (24, 25), and chemokine production
is induced in mice given T. gondii extract
(15). Despite these observations, the role of the defined
DC subsets during the course of an acute bacterial infection in vivo
has not been evaluated. The present study characterizes the
localization and function of CD8+,
CD8-CD4+, and
CD8-CD4- subsets of
CD11c+ splenic DC in response to virulent
Salmonella typhimurium acquired by the oral route.
Immunohistochemical analyses reveal differential temporal and
spatial involvement of CD4+ and
CD8+ subsets of CD11c+
splenocytes associated with progression of the infection. Furthermore,
ex vivo FACS analysis of DC subsets quantifies the response
within CD8+,
CD8-CD4+, and
CD8-CD4- subsets
during infection and demonstrates subset-specific changes in
cytokine profiles as part of the DC response. These phenomena occurring
during acute S. typhimurium infection shed new light on
possible differential in vivo functions for these recently defined DC
subsets.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
C57BL/6 mice and
TAP1/2-microglobulin
(2m) double knockout mice
(TAP1-/-2m-/-)
on a C57BL/6 background (26) were used between 8 and 12 wk
of age. All mice were bred and housed in the animal facilities at Lund
University and were provided with food and water ad libitum.
Bacterial strains and infection of mice
An 
asdA1 derivative of the SR-11 variant S.
typhimurium 
4666 (27) was made by transducing this
allele linked to zhf-4::Tn10 from S.
typhimurium 3520 as described (28). 4666 and
the asd derivative of Escherichia coli DH5
called E. coli 6212 (29) were used in these
studies. The strains carried the asd+
plasmid pYA3259 encoding OVA or an OVA-green-fluorescent protein fusion
construct.4 Bacteria
were cultured from frozen glycerol stocks overnight in Luria-Bertani
broth at 37°C with shaking. The bacterial concentration was estimated
spectrophotometrically, bacteria were washed and resuspended in sterile
PBS. Mice were inoculated intragastrically with 0.1 ml 1% sodium
bicarbonate followed 15–20 min later with bacteria in a volume of 0.2
ml. The actual bacterial dose given in each experiment was determined
by plating serial dilutions of the bacteria used for infections on
Luria-Bertani agar plates. The dose of bacteria used in the studies
ranged from 5 x 108 to 1 x
1010 bacteria/mouse for both S.
typhimurium and E. coli. No observable dose-related
difference in experimental outcome was apparent within this dose range
for either bacterium. Infections with this dose of S.
typhimurium resulted in a mild infection up to 7 days post
administration; mice showed no overt signs of infection, no deaths
occurred, and spleens showed little, if any, sign of enlargement on
sacrifice. Where stated, mice were infected with 1 x
105 bacteria/mouse to examine the effects on
animals given a low bacterial dose (
10,000 times lower than the
standard inoculum).
On sacrifice, spleens were removed aseptically, and the number of viable bacteria in the spleen of each animal was determined. A single-cell suspension of splenocytes was prepared as described below. Serial dilutions of splenocytes were subsequently plated on Luria-Bertani agar plates, and the number of colonies was determined. The total number of bacteria recovered per spleen was then calculated based on the total splenocyte count.
Monoclonal Abs
mAbs from the hybridomas GK1.5 (anti-CD4),
YTS169.4 (anti-CD8), 53.5.81 (anti-CD8), 145.2C11
(anti-CD3), N418 (anti-CD11c), 2.4G2 (anti-FcRII/III),
RA3.6B2 (anti-B220), M5/114 (anti-MHC-II), and C17.8
(anti-IL-12p40) (30) were used. mAbs HL3 (anti-CD11c)
and XMG1.2 (anti-IFN-) were from PharMingen (San Diego, CA). XT22
(anti-TNF-) was from Nordic BioSite (Stockholm, Sweden).
The mAbs R3-34 (rat IgG1), R35-95 (rat IgG2a), and A95-1
(rat IgG2b) (all from PharMingen) were used as isotype controls. In
immunohistochemistry studies of surface phenotype, primary mAbs were
applied unconjugated. mAbs used in flow cytometry were either directly
conjugated with PE, FITC, or allophycocyanin or were used biotinylated,
as described below.
Immunohistochemistry
Spleens taken from naive mice and at 4 h, 24 h, 48 h, 5 days and 7 days post-bacterial infection were laterally dissected into appropriately sized pieces, mounted in Tissue-Tek OCT (Sakura, Holland) and snap frozen in liquid nitrogen. Sections of spleen samples (7 µm) were cut, mounted on SuperFrost Plus slides (Menzel-Glaser, Freiburg, Germany), and air-dried. Detection of surface markers was conducted by a standard HRP method. Sections were fixed in ice-cold acetone, rehydrated in TBS, and blocked with 20% normal goat serum in TBS for 30 min. Primary mAbs were applied at 5–10 µg/ml in 20% normal goat serum in TBS for 1 h. Primary mAbs were detected with biotinylated F(ab')2 of mouse anti-rat Ig or with biotinylated goat anti-Armenian hamster IgG (both from Jackson ImmunoResearch Laboratories, West Grove, PA) followed by VectorStain Elite ABC kit (Vector Laboratories, Burlingame, CA). Staining was visualized using a diaminobenzidine (DAB) kit (Vector Laboratories). Sections were counterstained with hematoxylin (BDH Merck, Darmstadt, Germany). Finally, sections were dehydrated, mounted under NeoMount (BDH Merck), and assessed microscopically. Positive staining was never observed when either primary mAbs was omitted or when isotype control mAbs were used. Assessment of staining was conducted on an Olympus BX60 microscope fitted with a CoolSnap Pro digital camera and ImagePro Plus software (Media Cybernetics, Silver Spring, MD). Final images were produced from Adobe Photoshop 5.0 (Adobe Systems, Mountain View, CA).
Splenocyte preparation
Splenocytes were prepared by incubating small pieces of spleens with 1.6 mg/ml collagenase type IV (Worthington Biochemical, Freehold, NJ) and 1 mg/ml DNase 1 (Worthington Biochemical) in serum-free, calcium-free HBSS (Life Technologies, Paisley, U.K.) at 37°C for 45 min. Digested preparations were disaggregated by pipetting to produce a single-cell suspension. The cells were then washed in HBSS, and erythrocytes were lysed by hypotonic shock. Preparations were filtered to remove debris and were resuspended in IMDM (Life Technologies) containing 10% FCS. The total viable splenocyte number was determined by trypan blue exclusion. Cell suspensions were used to determine the bacterial load (as described above) and for flow cytometry.
Flow cytometry
To detect surface molecule expression, splenocytes were first
washed with wash buffer (HBSS containing 3% FCS, 1 mM EDTA, and 10 mM
HEPES). All subsequent steps were conducted in this buffer on ice.
Samples were first blocked with anti-FcRII/III mAb.
PE-conjugated anti-CD11c and 7-aminoactinomycin D (7AAD, Sigma)
were included in all subsequent stainings. For subset definition,
FITC-conjugated anti-CD4, anti-CD8, or anti-CD11b was
applied in conjunction with biotinylated anti-MHC-II followed by
streptavidin-allophycocyanin (PharMingen) for 30 min. In some samples,
to confirm the integrity of the detection and acquisition,
allophycocyanin conjugates of anti-CD4 or anti-CD8 were
applied in conjunction with FITC-conjugated anti-MHC-II. Samples
were then analyzed by four-color flow cytometry.
Intracellular cytokines were analyzed in splenocyte preparations that were first treated with 5 µg/ml brefeldin A (Sigma) for 5 h at 37°C in Ultra-Low Cluster 24-well tissue culture plates (Costar Corning, Cambridge, MA). Cells were washed and stained to detect surface molecules as described above. Anti-CD11c-PE and 7AAD were included in all stainings. Cells were then fixed with 2% paraformaldehyde in PBS and washed in permeabilization buffer (HBSS containing 0.5% saponin and 0.5% BSA (Sigma)). Intracellular cytokines were detected by addition of biotinylated mAbs followed by streptavidin-allophycocyanin diluted in permeabilization buffer. After final washes, cells were resuspended in wash buffer and analyzed by four-color flow cytometry. In some samples, the fluorochromes used to detect surface markers and cytokines were reversed to confirm the integrity of the detection and acquisition.
All samples were acquired using a FACSCalibur flow cytometer (Becton Dickinson, Mountain View, CA) and were analyzed using CellQuest software (Becton Dickinson). In surface phenotype studies at least 15,000 viable CD11c+ events were collected per sample. For intracellular cytokine analysis at least 10,000 viable CD11c+ events were collected per sample.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To evaluate the role of splenic DC subsets in the immune response
to acute bacterial infection, mice were orally challenged with either
S. typhimurium or E. coli. At 4 h, 24
h, 48 h, 5 days, or 7 days post-infection, the localization and
distribution of splenic DC subsets were analyzed. The bacterial load in
the spleen of each individual was also determined at the time of
sacrifice. At 48 h postinfection, the mean bacterial load in
Salmonella-challenged animals was
<103 total bacteria/spleen. This increased at
day 5 to a mean value of 104 total
bacteria/spleen. No bacteria were recovered from E.
coli-challenged animals at any time point. By
immunohistochemistry, Salmonella infection was associated
with a visually more "open" splenic architecture. This was apparent
at 48 h postinfection and was striking after 5 days. It was common
to all Salmonella-infected mice and was likely the result of
massive erythrocyte influx (our unpublished observations).
Animals challenged with E. coli exhibited no such changes
and, without exception, appeared directly comparable to naive mice by
immunohistochemical analyses.
Examination of serial sections revealed a population of cells located
at the margins of the white pulp that stained positive for both CD4 and
CD11c expression. Cells expressing CD4 and CD11c were particularly
prominent at the margins bordering the B cell areas (Fig. 1
, c and e). These
cells, however, did not stain positive for CD3 (Fig. 1a),
CD8, or CD8 (data not shown) and did not appear to have the
lymphocytic morphology observed on adjacent CD3+
cells. The presence of this population was confirmed at each time point
up to 48 h postinfection; no differences in the distribution were
apparent during the first 48 h in mice challenged with either
E. coli or Salmonella compared with naive animals
(Fig. 1 and data not shown). CD11c+ cells lacking
expression of CD8 and CD4 (i.e.,
CD8-CD4-
double-negative cells) did not appear to be significantly represented
within this area, although their presence could not be completely
excluded.
|
, d and f). This was not due to an
inability to detect
CD4+CD11c+ cells at this
stage of infection, in that such cells were clearly apparent within the
B cell areas themselves (Fig. 1, d and f). Few
CD3+CD4+ T cells (Fig. 1, b and d) or
CD4+CD11c+ cells (Fig. 1, d and f) were detected within the red pulp of
Salmonella-infected animals at this time point. The observed
altered distribution of
CD4+CD11c+ cells was
apparent only in Salmonella-infected mice; i.e., mice
challenged with E. coli demonstrated none of the above
phenotypic changes, appearing essentially like naive animals (data not
shown). The Salmonella-associated changes were also observed
in mice given a 10,000-fold lower dose of Salmonella
(105 CFU; n = 3; data not shown).
These data demonstrate the predominant association of CD4-expressing
CD11c+ cells with the marginal zones of B cell
areas in naive and E. coli-infected mice. Furthermore, this
DC subset is redistributed during Salmonella infection.
Redistribution of red pulp CD8+CD11c+
splenocytes during Salmonella infection
CD11c+ cells expressing CD8 were
identifiable to some extent in the red pulp of all
Salmonella-infected animals up to 48 h postinfection;
i.e., CD8 staining colocalizing with CD11c staining in excess of
CD8 staining was apparent in serial sections. During the first
48 h of infection, the distribution of
CD8+CD11c+ cells in the
red pulp of Salmonella-infected mice (data not shown) was
indistinguishable from naive or E. coli-infected mice (Fig. 2, d, f, and h).
However, a remarkable alteration in the distribution of CD8 and
CD11c-expressing cells, but not CD8+ cells,
was apparent throughout the tissue at day 5 of Salmonella
infection (Fig. 2). This was most obvious within the red pulp, where
large numbers of CD8+ cells were visible
compared with infrequent CD3+ and
CD8+ cells (Fig. 2, a, c, and
e). Moreover, the CD8 staining in the red pulp
colocalized with an equally distinct alteration in the distribution of
cells staining positive for CD11c (Fig. 2, e and
g). Although coincident CD8CD8-staining cells were
present in the white pulp, these cells were very scarce in the red pulp
of these samples (Fig. 2, c and e). The observed
alteration in red pulp cells that stained positive for CD8 and CD11c
but not for CD3 or CD8 was never observed at time points before 5
days postinfection. Furthermore, the modulation of the
CD8+CD11c+ cell
population was observed in all S. typhimurium-infected
animals examined at this time point (n = 6) and was not
observed in E. coli-challenged animals (n =
3). The latter exhibited CD8+ and
CD11c+ distributions similar to that of naive
mice (n = 6). Animals were also examined 7 days
postinfection. The redistribution of red pulp CD8 and CD11c
compartments were even more striking at this time and was observed in
all Salmonella-infected animals (n = 4; data
not shown).
|
staining
alone (Fig. 2, e and g). This suggested that
other population(s) of CD11c+ cells lacking
CD8 expression were present. However, CD4 staining was infrequent in
the red pulp and colocalized with CD3 staining at this stage of
infection (Fig. 1, b and d). This suggests the
presence of numerous CD11c+ cells that express
neither CD8 nor CD4, i.e.,
CD8-CD4-
CD11c+ cells, in the red pulp 5 days
post-Salmonella infection. Indeed, this qualitative
observation was confirmed by quantitating the absolute number of the DC
subsets during infection (see below). The alterations
apparent in both the CD8+ and
CD8-CD4-
CD11c+ populations were observed when mice were
orally immunized with a lower dose of Salmonella
(105 CFU; n = 3; data not shown)
but were never apparent in either naive mice or animals challenged with
E. coli.
Immunohistochemistry on spleen sections from orally infected
TAP1-/-2m-/-
mice, which have exceptionally low numbers of splenic
CD8+ T cells (26), confirmed the
above observations; i.e., at day 5 of Salmonella infection,
both CD8+ and CD11c+
cells were redistributed in the red pulp in a manner similar to that
seen in C57BL/6 mice (Fig. 3). These
changes were, as in wild-type animals, independent of CD3 (Fig. 3b) and CD8 staining (not shown), strongly indicating
that oral Salmonella infection induces a significant in vivo
redistribution of splenic CD8+ and
CD8-CD4- DC
populations. Together these data show that
CD8+ and
CD8-CD4-CD11c+
cell frequencies in the red pulp increase dramatically 5 days after
Salmonella infection.
|
To quantitatively analyze alterations in DC populations during
acute Salmonella infection, flow cytometry analysis was
performed on splenocytes from bacterially challenged mice at various
time points after oral infection. Splenic DC, defined as cells
expressing high levels of both CD11c and MHC-II (Fig. 4a), showed only a slight
increase as a percentage of total splenocytes 48 h after
Salmonella infection (Fig. 4b). At 5 days
postinfection, however, the percent of splenic DC was significantly
elevated in Salmonella-infected mice (Fig. 4b).
No significant increase in DC as a percentage of total splenocytes was
observed in animals challenged with E. coli at any time
point, and E. coli-challenged mice did not differ
significantly from naive mice (n = 6).
|
+ or
CD8-CD4+ subsets among
total DC was observed in Salmonella-challenged animals until
5 days postinfection (Fig. 4c). At this time, a decreased
proportion of CD8-CD4+
and a coincident rise in CD8+ DC was apparent
compared with E. coli-challenged mice (Fig. 4c).
By calculating the number of DC that stained with neither anti-CD4
nor anti-CD8 mAbs, a population of
CD8-CD4-
double-negative DC was also analyzed in each sample. These data
revealed that the CD8+,
CD8-CD4+, and
CD8-CD4- subsets
constituted 25.4 ± 2.6, 49.2 ± 4.4, and 24.0 ± 4.2%
of total DC, respectively, in naive animals (n = 6).
Splenocyte samples from Salmonella-infected mice analyzed in
this manner did not reveal a significant alteration in the percentage
of CD8-CD4- DC
compared with either E. coli-challenged animals (Fig. 4c) or naive mice. Data obtained in independent flow
cytometry analyses that directly quantitated the
CD8-CD4- DC subset
from the same samples confirmed these observations (data not shown).
Finally, the percentage of total DC and of
CD8+,
CD8-CD4+, and
CD8-CD4- DC subsets
from naive mice (n = 9; data not shown) did not differ
significantly from the values obtained for E. coli-infected
mice at all time points (n = 6), demonstrating that the
observed quantitative alterations in DC subset numbers were associated
with Salmonella infection.
The total number of splenocytes in mice infected 2 or 5 days earlier
with Salmonella was typically 1–1.2 times that of the other
groups. The absolute number of splenic DC of each subset was
quantitated for each animal based on the total number of viable
splenocytes (Table I). These calculations
revealed that the absolute number of
CD4+CD11c+ cells remained
constant during Salmonella infection. This was despite their
redistribution in situ (Fig. 1) and reduction as a percentage of total
splenic DC at 5 days postinfection (Fig. 4c). The absolute
number of both the CD8+ and
CD8-CD4- DC
subpopulations was significantly increased at day 5 after
Salmonella infection. In contrast, no changes were observed
in E. coli-challenged mice (Table I), which had absolute
numbers of each DC subset that did not differ significantly from those
of naive mice.
|
2m-/-
double-knockout mice confirmed these data. Thus, no significant change
in the absolute number of
CD8-CD4+ DC was seen at
day 5 postinfection (data not shown). In contrast, the absolute number
of CD8+ DC was significantly increased at this
time point (mean values of 12.5 ± 3.6 x
105 and 6.8 ± 2.8 x
105 for Salmonella-infected and naive
TAP1-/-2m-/-
mice, respectively. p = 0.024). Similarly,
CD8-CD4- DC were also
significantly increased at day 5 (mean values of 7.9 ± 0.5
x 105 and 4.5 ± 0.4 x
105 for Salmonella-infected and naive
TAP1-/-2m-/-
mice, respectively. p = 0.001). Together, these data
provide quantitative confirmation of the immunohistochemistry
observations (Figs. 2 and 3) and support the subset-specific dynamics
of DC populations during Salmonella infection. Quantitative changes in cytokine production among DC subsets during Salmonella infection
Initial immunohistochemical analyses of spleen sections
investigated DC cytokine production during Salmonella
infection. These data showed increased TNF- and IFN- expression
in the red pulp as early as day 2 post-Salmonella challenge,
and many TNF-- and IFN--positive cells occurred in clusters in
the red pulp at day 5. Few cells staining positive for either of these
cytokines were apparent in the red pulp of naive mice or mice
challenged 2 or 5 days previously with E. coli. Furthermore,
expression of TNF- or IFN- was not observed within the white pulp
of any samples, nor did it correspond in location to the
CD4+CD11c+ subset (data not
shown).
These analyses were, however, unable to clarify cytokine expression by
DC during Salmonella infection. Thus, intracellular cytokine
staining and flow cytometry analyses were performed ex vivo. For these
analyses, splenic DC were defined as viable cells expressing a high
level of CD11c (Fig. 5a). More
than 95% of the cells within the gated CD11chigh
population also expressed high levels of MHC-II (Fig. 5a), ensuring that cytokine productionby DC
was being analyzed. TNF- was produced by
CD11chighcells, and essentially
all of the cytokine detected was producedby the
CD11chighMHC-IIhigh
population (Fig. 5a). CD8+ and
CD8-CD4+ DC subsets
were further analyzed for cytokine production.
|
+ and
CD8-CD4+ DC produced
TNF- at day 2 of Salmonella infection (Table II). CD8+ DC
TNF- expression increased postinfection such that the absolute
number of CD8+CD11c+
cells producing TNF- was significantly greater in S.
typhimurium-infected mice than in E. coli-challenged
animals at day 5 (Fig. 5b and Table II). In the
CD4+CD11c+ DC subset,
however, the trend was the opposite at this time point; i.e., the
absolute number of CD4+ DC from
Salmonella-infected animals staining positive for TNF-
was reduced at day 5 to a level below that seen in E.
coli-challenged mice (Table II).
|
+ DC
producing IL-12p40 was apparent at day 2 of Salmonella
infection (Table III). At 5 days
postinfection, however, reduced numbers of IL-12p40-expressing
CD4+ and CD8+ DC were
apparent in mice that received Salmonella. In the
CD4+ DC subset, this reduction was significant
compared with E. coli-infected animals (Table III).
Additional studies addressed whether the peak of IL-12 production in
Salmonella-challenged mice occurred before 48 h.
However, no evidence was found for a significant increase in the
numbers of IL12p40-expressing cells of either DC subset 18–24 h
postinfection (n = 3; data not shown). Furthermore,
IL-10 expression was not detected in either CD4+
or CD8+ DC subsets during the response to
either E. coli (48 h, n = 3; 5 days,
n = 3) or S. typhimurium (48 h,
n = 3; 5 days, n = 5; data not shown)
in experiments in which IL-10 was detectable in parallel samples of
splenocytes treated with ionomycin and PMA. Thus, IL-10 production by
splenic DC does not appear to contribute to the reduced DC IL-12
production in the later stages of infection. These results demonstrate
subset-specific modulations of DC cytokine expression in response to
Salmonella infection.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
+,
CD8-CD4+, and
CD8-CD4- DC subsets
during the response to oral S. typhimurium infection. The in
vivo localization of S. typhimurium in
CD11c+ cells (34) and the activation
of DC during Salmonella infection (our unpublished
observations) underscore the importance of elucidating the role of DC
subsets during the host response to this bacterium.
Our data demonstrate in situ that CD11c+ cells
expressing CD4 are associated with the marginal zones in normal mice.
No CD8 expression and little evidence of
CD8-CD4- DC was
observed in this area. CD4+ DC were also present
within the B cell areas themselves. Five days
post-Salmonella infection, however,
CD4+ DC disappear from the marginal zones but not
the B cell areas. These data, combined with unchanged absolute numbers
of CD4+ DC during Salmonella
infection, support a redistribution of the CD4+
subset rather than, e.g., selective death of these cells. This is in
contrast to an overt reduction in splenic CD11c+
cells by apoptosis that occurs in vivo as a result of LPS
administration (24, 25). Furthermore,
Salmonella-induced apoptosis of infected cells (35, 36) is not likely responsible for the loss of
CD4+ DC from the marginal zones for two reasons.
First, such a mechanism cannot readily explain the loss of one DC
subset from a particular splenic location whereas other DC populations,
such as the CD8+ and
CD8-CD4- subsets,
increase during infection. Second, S. typhimurium shows no
apparent preference for association with either
CD4+ or CD8+ DC subsets
in vivo (our unpublished observations). Finally, it is unlikely that
down-regulation of CD4 expression is involved in the loss of
CD4+ DC from the marginal zone. Although CD4
expression on splenic DC is reduced upon in vitro culture (10, 37) or after systemic LPS treatment (37), the
presence of CD4+ DC within the B cell follicles
and the lack of a reduction in
CD8-CD4+ DC argues
against CD4 down-regulation as an explanation for the change in this
population during Salmonella infection. Whether
CD4+ DC migrate from the marginal zones into the
follicles, die at the B cell margins and expand elsewhere, or immigrate
from other site(s) during infection remains to be determined.
DC were present at a high frequency within splenic T cell areas at all
times (see Figs. 2 and 3). Thus, determining distinct DC
redistributions to or within T cell areas during the course of
Salmonella infection was difficult. This was also the case
when CD8+ DC were analyzed in
TAP1-/-2m-/-
mice. Instead, the immunohistochemical data show that both
CD8+ and
CD8-CD4- subsets are
present within the red pulp and are not limited to T cell areas.
Moreover, Salmonella infection results in a striking
increase in the frequency of both CD8+ and
CD8-CD4- DC within the
red pulp after 5–7 days. This finding was confirmed in
TAP1-/-2m-/-
mice, which are essentially devoid of CD8+ T
cells (26).
No specific increase in DC subsets in response to bacterial infection
has previously been reported. However, systemic administration of LPS
or a soluble T. gondii extract results in apparent
redistribution of CD11c+ or
DEC205+ cells from the periarterial lymphoid
sheath to the T cell areas (12, 13, 15, 20, 24). These
changes occur rapidly, within hours after treatment. In contrast, oral
infection with S. typhimurium did not result in a noticeable
redistribution of splenic DC until 
48 h post challenge. A difference
in the kinetics of Ag delivery and in the overall antigenic load in the
spleens of animals given LPS or protein extract i.v. compared with oral
Salmonella may underlie this distinction. The
Salmonella-dependent alteration of DC distribution within
the splenic red pulp may instead be an aspect of the early DC response
to Salmonella infection. Consequently, these red pulp DC may
have infection-associated functions specific for this area early during
infection. Such functions may include Salmonella
internalization in vivo (our unpublished observations), or uptake and
cross-presentation of Salmonella Ags derived from apoptotic
and/or necrotic cells (36). Studies using systemically
administered protein Ag showed that DC presenting the Ag appear first
within the outer periarteriolar lymphoid sheath and are detectable in
splenic T cell areas within 24 h (20). However,
specific T cell responses to oral Salmonella are barely
detectable after 1 wk and may require 3–4 wk to fully develop
(38). This may suggest that DC in the red pulp will
further migrate to interact with T cells later during
Salmonella infection.
Quantitative changes in DC subsets were also apparent in response to
Salmonella infection. That is, although no change in
CD8-CD4+ DC numbers
occurred during infection, increases in the
CD8+ and
CD8-CD4- subsets were
significant after 5 days. These increases are likely consequences of DC
migration to the spleen during infection, due to the role of the spleen
in initiating specific immunity and as a site of Salmonella
replication. Similarly, splenic DC numbers increase after
administration of soluble T. gondii extract, which was
suggested to result from an influx of blood-borne cells
(13). LPS also causes the release of DC from peripheral
sites such as the skin and intestinal lamina propria via a mechanism
involving TNF- (39, 40). It is therefore conceivable
that oral infection with Salmonella may initiate similar
responses through both bacterial- and host-derived products. Because
the present data show that Salmonella infection induces
quantitative increases in splenic DC in a subset-specific manner, the
mechanism(s) underlying the observed changes must differentially
influence the subsets. It is possible that subset- and
location-restricted expression of chemokine receptor/ligand pairs may
play a role in differential recruitment. Indeed, differential
expression of mRNAs for chemokine receptors within both splenic- and
Peyer’s patch-derived DC subsets has been demonstrated
(21). Such mechanisms may contribute to DC recruitment in
response to Salmonella infection.
Our data also show subset-specific, Salmonella-associated
changes in DC cytokine production. That is, transiently increased
numbers of TNF--producing CD4+ DC at day 2
postinfection waned considerably at day 5 concomitant with the possible
redistribution of these cells to the B cell areas. In contrast,
increased numbers of TNF--producing CD8+ DC
at day 2 further increased at day 5 of infection. However, in situ data
showed that TNF- production at these times was not detected in the
white pulp and did not colocalize with CD8+ DC
in the red pulp. Rather, it is primarily associated with neutrophils
(our unpublished observations), cells critical in the early response to
Salmonella (41). Despite significant increases
in TNF--producing CD8+ DC in
Salmonella infection, the absolute number of these cells is
relatively low. Together these observations suggest that TNF-
production by splenic DC may not primarily be involved in controlling
the infection. Instead, a function of TNF- produced by DC in
response to Salmonella both in vivo, as shown here, and in
vitro (42) may be to induce DC maturation
(43, 44, 45). Thus, red pulp CD8+ DC
may produce and respond to TNF- in an autocrine (46) or
paracrine manner at early times after infection. This may lead to
maturation and subsequently migration of DC into the T cell areas to
initiate specific immunity.
IFN- is required for effective protection against
Salmonella (47). At 2 days
post-Salmonella challenge, IFN--expressing splenocytes
were apparent by immunohistochemistry, but DC producing IFN- were
not detected by ex vivo flow cytometry (our unpublished observations).
It is likely that macrophages and NK cells are a major source of
IFN- at this time during infection (47). However, by
day 5 after Salmonella infection, IFN--expressing
CD8+ and
CD8-CD4+ DC were
present in six of six mice (data not shown). Although the frequency of
these cells was low (1% of the gated population), the absence of
IFN--positive DC in naive or E. coli-challenged animals
supports the idea that DC IFN- production is elicited by
Salmonella infection. The functional significance of this,
however, remains to be determined.
IFN- production can be regulated by IL-12 (17, 48).
However, significantly increased IL-12p40 production by either
CD8+ or
CD8-CD4+ DC was not
observed in the time frame of Salmonella infection studied
here. This may seem surprising in light of previous findings that LPS
or T. gondii extract induce high levels of DC IL-12
production (12, 13, 15, 16). It is possible that IL-12
production by splenic DC in Salmonella infection may occur
at a time other than those analyzed (i.e., <18 h or between 2 and 5
days postinfection). An early peak of transient IL-12 production after
Salmonella infection would be consistent with that observed
in mice injected with T. gondii extract, which peaks <6 h
post administration (12, 13, 15, 16). However, as
discussed above, DC responses during oral infection with
Salmonella clearly progress with kinetics different from
those for i.v. applied soluble stimuli and a slower, less pronounced
IL-12 response may be appropriate.
The lack of IL-12 detected in vivo is not due to an inherent inability
of DC to produce this cytokine in response to Salmonella.
Indeed, Salmonella readily induces IL-12 production by DC in
vitro (42). This work, and that of others (13, 15), suggests that IL-12 production results from direct
interaction between a microbial component and DC, possibly LPS in the
case of Salmonella (42). The lack of
significant DC IL-12 production is likely a reflection of the limited
numbers of Salmonella present in the spleen during the early
stages of in vivo infection. For example, <1% of splenic DC associate
with Salmonella 4 h after oral challenge, and extremely
low numbers of splenic DC associated with bacteria are apparent at
later time points (our unpublished observations). Thus, IL-12
production by splenic DC in response to Salmonella infection
may be numerically, as well as temporally, restricted. Because we were
unable to detect DC production of IL-10 in response to
Salmonella infection, it is unlikely that DC-derived IL-10
abrogated IL-12 production (49). Finally, equivalent,
albeit low, numbers of DC expressing TNF- and IL-12p40 were detected
in naive and E. coli-challenged mice, suggesting that some
constitutive expression of these cytokines by DC may occur.
Although DC have a unique role in the initiation of immunity, these
cells function as part of a response that mobilizes multiple cell types
to combat Salmonella during the early stages of infection
(47). Ultimately, the extent and nature of DC involvement,
including the role of DC subsets, may determine the exact nature of the
specific immune response. Our data suggest that cells other than DC may
likely be the major source of cytokines (i.e., TNF-, IFN-, and
IL-12) important in controlling Salmonella replication
during the early stages of infection. In vivo, DC responses to
Salmonella infection are likely to be spread over a
relatively long time period. However, it is clear that DC respond to
Salmonella infection in a subset-specific fashion. Such
temporal and spatial coordination of DC involvement may optimize
functional DC responses in anti-Salmonella immunity.
|
Acknowledgments |
|---|
6212, S.
typhimurium 3520 and plasmid pYA3259. Dr. Mikael Rhen
(Karolinska Institute, Stockholm, Sweden) generously provided S.
typhimurium 4666. We also thank Dr. Giorgio Trinchieri
(Wistar Institute of Anatomy and Biology, Philadelphia, PA) for
providing hybridoma C17.8. |
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Mary Jo Wick, Department of Cell and Molecular Biology, Section for Immunology, Lund University, BMC I-13, 221 84, Lund, Sweden. E-mail address: Mary_Jo.Wick{at}immuno.lu.se
3 Abbreviations used in this paper: DC, dendritic cells; 2m, 2-microglobulin; 7AAD, 7-aminoactinomycin D; DAB, diaminobenzidine.
4 U. Yrlid, M. Svensson, A. Håkansson, B. Chambers, H.-G. Ljunggren, and M. J. Wick. In vivo activation of dendritic cells and T cells during Salmonella enterica Serovar Typhinuvium infection. Submitted for publication.
Received for publication December 26, 2000. Accepted for publication March 22, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- and CD8+ dendritic cells are generated from CD4low lymphoid-committed precursors. Blood 96:2511.+ and CD8- subclasses of dendritic cells direct the development of distinct T helper cells in vivo. J. Exp. Med. 189:587.+ dendritic cells. Nat. Immunol. 1:83.[Medline]
production by CD8+ lymphoid dendritic cells. J. Exp. Med. 189:1981., MIP-3, and secondary lymphoid organ chemokine. J. Exp. Med. 191:1381.2-microglobulin double mutant mice. Int. Immunol. 7:975. production and lethality in lipopolysaccharide-induced shock in mice. Eur. J. Immunol. 25:672.[Medline]
. J. Exp. Med. 179:1109., and calcium ionophore under serum-free conditions promote rapid dendritic cell-like differentiation in CD14+ monocytes through distinct pathways that activate NF-
B. J. Immunol. 165:3647. production and type 1 cytokine responses. Immunity 4:471.[Medline]
This article has been cited by other articles:
|
|
 |
|
  A. Maroof and P. M. Kaye Temporal Regulation of Interleukin-12p70 (IL-12p70) and IL-12-Related Cytokines in Splenic Dendritic Cell Subsets during Leishmania donovani Infection Infect. Immun., January 1, 2008; 76(1): 239 - 249. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Rydstrom and M. J. Wick Monocyte Recruitment, Activation, and Function in the Gut-Associated Lymphoid Tissue during Oral Salmonella Infection J. Immunol., May 1, 2007; 178(9): 5789 - 5801. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Raine, P. Zaccone, P. Mastroeni, and A. Cooke Salmonella typhimurium Infection in Nonobese Diabetic Mice Generates Immunomodulatory Dendritic Cells Able to Prevent Type 1 Diabetes J. Immunol., August 15, 2006; 177(4): 2224 - 2233. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Ato, A. Maroof, S. Zubairi, H. Nakano, T. Kakiuchi, and P. M. Kaye Loss of Dendritic Cell Migration and Impaired Resistance to Leishmania donovani Infection in Mice Deficient in CCL19 and CCL21 J. Immunol., May 1, 2006; 176(9): 5486 - 5493. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Sundquist and M. J. Wick TNF-{alpha}-Dependent and -Independent Maturation of Dendritic Cells and Recruited CD11cintCD11b+ Cells during Oral Salmonella Infection J. Immunol., September 1, 2005; 175(5): 3287 - 3298. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Rosales-Reyes, C. Alpuche-Aranda, M. d. l. L. Ramirez-Aguilar, A. D. Castro-Eguiluz, and V. Ortiz-Navarrete Survival of Salmonella enterica Serovar Typhimurium within Late Endosomal-Lysosomal Compartments of B Lymphocytes Is Associated with the Inability To Use the Vacuolar Alternative Major Histocompatibility Complex Class I Antigen-Processing Pathway Infect. Immun., July 1, 2005; 73(7): 3937 - 3944. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. Kirby, M. Sundquist, and M. J. Wick In Vivo Compartmentalization of Functionally Distinct, Rapidly Responsive Antigen-Specific T-Cell Populations in DNA-Immunized or Salmonella enterica Serovar Typhimurium-Infected Mice Infect. Immun., November 1, 2004; 72(11): 6390 - 6400. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Leisewitz, K. A. Rockett, B. Gumede, M. Jones, B. Urban, and D. P. Kwiatkowski Response of the Splenic Dendritic Cell Population to Malaria Infection Infect. Immun., July 1, 2004; 72(7): 4233 - 4239. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. C. Alaniz, S. Sandall, E. K. Thomas, and C. B. Wilson Increased Dendritic Cell Numbers Impair Protective Immunity to Intracellular Bacteria Despite Augmenting Antigen-Specific CD8+ T Lymphocyte Responses J. Immunol., March 15, 2004; 172(6): 3725 - 3735. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Kalupahana, A. R. Emilianus, D. Maskell, and B. Blacklaws Salmonella enterica Serovar Typhimurium Expressing Mutant Lipid A with Decreased Endotoxicity Causes Maturation of Murine Dendritic Cells Infect. Immun., November 1, 2003; 71(11): 6132 - 6140. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-N. Tournier, A. Q. Hellmann, G. Lesca, A. Jouan, E. Drouet, and J. Mathieu Fever-like thermal conditions regulate the activation of maturing dendritic cells J. Leukoc. Biol., April 1, 2003; 73(4): 493 - 501. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. Kirby, U. Yrlid, and M. J. Wick The Innate Immune Response Differs in Primary and Secondary Salmonella Infection J. Immunol., October 15, 2002; 169(8): 4450 - 4459. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Yrlid and M. J. Wick Antigen Presentation Capacity and Cytokine Production by Murine Splenic Dendritic Cell Subsets upon Salmonella Encounter J. Immunol., July 1, 2002; 169(1): 108 - 116. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Jiao, R. Lo-Man, P. Guermonprez, L. Fiette, E. Deriaud, S. Burgaud, B. Gicquel, N. Winter, and C. Leclerc Dendritic Cells Are Host Cells for Mycobacteria In Vivo That Trigger Innate and Acquired Immunity J. Immunol., February 1, 2002; 168(3): 1294 - 1301. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Yrlid, M. Svensson, A. Hakansson, B. J. Chambers, H.-G. Ljunggren, and M. J. Wick In Vivo Activation of Dendritic Cells and T Cells during Salmonella enterica Serovar Typhimurium Infection Infect. Immun., September 1, 2001; 69(9): 5726 - 5735. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or
Salmonella-infected (
) mice. d, Day.
c, Percent of splenic DC expressing CD4 (
) or CD8