Rapid Accumulation of CD14+CD11c+ Dendritic Cells in Gut Mucosa of Celiac Disease after in vivo Gluten Challenge
et al. (2012) Rapid Accumulation of CD14+CD11c+ Dendritic Cells in Gut Mucosa of Celiac
Disease after in vivo Gluten Challenge. PLoS ONE 7(3): e33556. doi:10.1371/journal.pone.0033556
+ + Rapid Accumulation of CD14 CD11c Dendritic Cells in Gut Mucosa of Celiac Disease after in vivo Gluten Challenge
Ann-Christin Rberg Beitnes 0
Melinda Ra ki 0
Margit Brottveit 0
Knut Erik Aslaksen Lundin 0
Lars Jahnsen 0
Ludvig Magne Sollid 0
Karol Sestak, Tulane University, United States of America
0 1 Centre for Immune Regulation and Department of Immunology, Oslo University Hospital - Rikshospitalet , Oslo , Norway , 2 Department of Medicine, Oslo University Hospital - Ulleva l and University of Oslo , Oslo , Norway , 3 Department of Medicine, Oslo University Hospital - Rikshospitalet , Oslo , Norway , 4 Centre for Immune Regulation and Department of Pathology, Oslo University Hospital - Rikshospitalet and University of Oslo , Oslo , Norway , 5 Centre for Immune Regulation and Department of Immunology, University of Oslo , Oslo , Norway
Background: Of antigen-presenting cells (APCs) expressing HLA-DQ molecules in the celiac disease (CD) lesion, CD11c+ dendritic cells (DCs) co-expressing the monocyte marker CD14 are increased, whereas other DC subsets (CD1c+ or CD103+) and CD163+CD11c2 macrophages are all decreased. It is unclear whether these changes result from chronic inflammation or whether they represent early events in the gluten response. We have addressed this in a model of in vivo gluten challenge. Methods: Treated HLA-DQ2+ CD patients (n = 12) and HLA-DQ2+ gluten-sensitive control subjects (n = 12) on a gluten-free diet (GFD) were orally challenged with gluten for three days. Duodenal biopsies obtained before and after gluten challenge were subjected to immunohistochemistry. Single cell digests of duodenal biopsies from healthy controls (n = 4), treated CD (n = 3) and untreated CD (n = 3) patients were analyzed by flow cytometry. Results: In treated CD patients, the gluten challenge increased the density of CD14+CD11c+ DCs, whereas the density of CD103+CD11c+ DCs and CD163+CD11c2 macrophages decreased, and the density of CD1c+CD11c+ DCs remained unchanged. Most CD14+CD11c+ DCs co-expressed CCR2. The density of neutrophils also increased in the challenged mucosa, but in most patients no architectural changes or increase of CD3+ intraepithelial lymphocytes (IELs) were found. In control tissue no significant changes were observed. Conclusions: Rapid accumulation of CD14+CD11c+ DCs is specific to CD and precedes changes in mucosal architecture, indicating that this DC subset may be directly involved in the immunopathology of the disease. The expression of CCR2 and CD14 on the accumulating CD11c+ DCs indicates that these cells are newly recruited monocytes.
Funding: This study was supported by grants from Oslo University Hospital Rikshospitalet and the Research Council of Norway. The funders had no role in
study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
. These authors contributed equally to this work.
Celiac disease (CD) is a chronic small intestinal inflammatory
condition caused by an inappropriate immune response to gluten
proteins of wheat, rye and barley. The condition is common with a
prevalence of about 1%. There is a strong association of CD with
certain HLA alleles. The majority of patients carry HLA-DQ2.5
(8095%) whereas most of the remaining patients carry
HLADQ8 [1,2]. The histopathology of CD is characterized by villous
blunting, crypt hyperplasia and increased number of CD3+
intraepithelial lymphocytes (IELs) . The current treatment for
CD is lifelong gluten-free diet (GFD).
Gluten-reactive T cells in the gut appear to play a central role in
the immunopathogenesis of CD, but how and where these T cells
get activated by interaction with antigen-presenting cells (APCs)
fronting them with gluten antigen are not well understood. The
priming of nave T cells is likely to take place in organized
lymphoid tissue whereas activation of effector T cells probably
takes place in the gut mucosa. In a previous study we found that
duodenal lamina propria HLA-DQ+ APCs could be divided into
CD11c+ dendritic cells (DCs) and CD68+CD11c2 macrophages
and that gluten challenged CD11c+ DCs isolated from celiac
lesions more efficiently activated gluten-reactive T cells than their
macrophage counterparts . More recently we further
subclassified the CD11c+ DC population into cells expressing either
CD103, CD1c or CD163. Co-staining experiments showed that
CD68+ and CD163+ cells were completely overlapping
populations. CD103+ and CD1c+ DCs were partly overlapping
populations, whereas the majority of CD11c+CD163+ DCs
coexpressed the monocyte marker CD14, suggesting that they were
derived from monocytes. Furthermore, we observed that the
density of CD14+CD11c+ DCs was increased while the density of
CD103+ DCs, CD1c+ DCs and CD163+CD11c2 macrophages
was decreased in the active celiac lesion . The differences
between untreated CD patients and healthy controls with respect
to mucosal APC populations suggest dynamic regulation of these
populations, but little is known about the kinetics of these changes
in relation to gluten exposure.
To address this issue we used a unique in vivo model in which the
density of mucosal APC subsets of treated CD patients and
glutensensitive control subjects where determined before and after a
three-day gluten challenge. We show that there are rapid dynamic
changes in the APC populations in the challenged duodenal
mucosa of CD patients, but not in gluten-sensitive controls.
Materials and Methods
The study was approved by the Regional Committee for
Medical Research Ethics in South-East Norway and the Privacy
Ombudsman for Research at Oslo University Hospital
Rikshospitalet (Oslo, Norway), and it is registered at http://
clinicaltrials.gov/ct2/results?term=NCT01100099. The study
complies with the Declaration of Helsinki. All the participants
gave their written informed consent.
We studied CD patients (n = 12; mean age 51 years, range 37
65, 7 females) and gluten-sensitive control subjects (n = 12; mean
age 45 years, range 2965, 12 females) who all were HLA-DQ2+.
The subjects have been described in detail elsewhere . Twelve
CD patients, diagnosed on the basis of typical histopathological
changes in duodenal mucosa  and from whom we had
cryopreserved biopsies available, as well as twelve randomly
selected gluten-sensitive control subjects were included. Both the
CD patients and the gluten-sensitive control subjects had been on
a strict GFD for at least 4 weeks prior to the study. The
glutensensitive participants had initiated their GFD without being
examined for CD by gastroendoscopy. Three had negative IgA
TG2 serology before commencing the GFD and nine had
unknown serology. Most of these subjects had experienced
symptoms, like abdominal discomfort and/or diarrhea, on a
gluten containing diet, and the symptoms improved on dietary
gluten elimination. All participants were challenged orally with
four slices (,160 g) of gluten-containing white bread every day for
three days. Duodenal biopsy specimens were obtained before
challenge and at day four. During the gluten challenge, symptoms
like bloating, diarrhea, constipation and satiety were observed in
six of twelve CD patients and in seven of twelve gluten-sensitive
control subjects. There were only four CD patients who
experienced histological changes after challenge; two patients
changed from Marsh 1 to 3a, one patient changed from Marsh 2
to 3a, and one patient changed from Marsh 0 to 3b. One patient
was unchanged Marsh 3a, whereas the seven other patients were
unchanged Marsh 0. Thus, there were no statistically significant
changes in Marsh grade after a three-day gluten challenge
amongst the CD patients. Biopsies from gluten-sensitive control
subjects were scored as Marsh 0 both before and after challenge in
all cases. The gluten-sensitive subjects were included as controls
because they adhered to a GFD and were willing to undergo a
gluten challenge. These subjects are highly unlikely to suffer from
CD as they did not react with appearances of HLA-DQ2-gluten
tetramer positive CD4+ T cells in the peripheral blood after the
gluten challenge .
In addition to the material obtained from individuals of the
challenge study, duodenal biopsies were obtained from three
treated CD and three untreated CD patients as well as from four
patients with normal histology who were examined with
gastroendoscopy as part of the routine diagnostic workup. Blood samples
were also obtained from two treated CD patients who were not
challenged with gluten. These subjects were included due to
limited material available from the participants of the challenge
study. Finally, CD14+ monocytes were isolated from buffy coats
obtained from two HLA-DQ2+ healthy individuals.
Multicolor immunofluorescence staining
Two biopsy specimens from each subject were oriented on thin
slices of carrot, embedded in Tissue Tek optimal cutting
temperature (O.C.T.) compound, snap frozen bed-side in liquid
nitrogen and stored at 270uC. H+E stained sections of both
specimens were evaluated and cryosections of the best oriented
tissue sample were cut in series at 4 mm and dried in room
temperature (RT) over night. The sections were then fixed with
acetone for 10 minutes, dried for 15 minutes, wrapped in
aluminium foil and stored at 220uC until use. Two- or
threecolor immunofluorescence staining was performed. To determine
the density and phenotype of HLA-DQ+ APCs, cryosections were
first incubated with the following combinations of mouse
monoclonal antibodies (mAbs) for 1 hour at RT: anti-HLA-DQ
(clone SPV-L3, IgG2a, 1.3 mg/mL, kind gift from H. Spits,
Amsterdam, Netherlands ) with anti-CD163 (clone RM3/1,
IgG1, 10 mg/mL, lot 818176, abcam, Cambridge, UK);
antiCD11c (clone CRB-p150/4G1, IgG2a, 5 mg/mL; Biosource,
Camarillo, CA) with either anti-CD103 (clone Ber-ACT8, IgG1,
1/200, kind gift from H. Du rkop, Berlin, Germany ), CD1c
(clone M241, IgG1, 5 mg/mL, Ancell Corporation, Bayport, MN),
CD14 (clone 18D11, IgG1, 0.3 mg/mL, kind gift from T. Espevik,
Trondheim, Norway ) or CD163 (clone RM3/1). Polyclonal
antibody specific for cytokeratin (rabbit anti-human IgG, 1/100,
H. Huitfeldt, Oslo, Norway ) was added to the primary mAb
mixtures to visualize the epithelium. The sections were briefly
rinsed with phosphate-buffered saline (PBS) and then incubated
with biotinylated goat anti-mouse IgG2a (2.5 mg/mL;
SouthernBiotech, Birmingham, AL) for 1.5 hour, followed by a
combination of Cy2-labeled streptavidin (1 mg/mL, Amersham
Biosciences, Buckinghamshire, UK) and Cy3-labeled goat
antimouse IgG1 (2.9 mg/mL; SouthernBiotech) for 30 minutes. Initial
testing showed that the staining intensity using anti-CD103 was
enhanced by prefixing the sections with
paraformaldehyde-lysineperiodate (PLP) for 10 minutes. In that case, incubation with
antiCD103 and anti-CD11c was followed by a combination of
FITClabeled goat anti-mouse IgG2a (20 mg/mL; SouthernBiotech) and
Cy3-labeled goat anti-mouse IgG1 for 30 minutes. Before
mounting, the sections were washed in Hoechst 33258,
pentahydrate (bis-benzimide) (1 mg/mL; Invitrogen, Paisley, UK) for
5 minutes to visualize cell nuclei. When anti-cytokeratin was
included, 7-amino-4-methylcoumarin-3-acetic acid
(AMCA)-labeled goat anti-rabbit IgG (1/101/20; Vector Laboratories,
Burlingame, CA) was added in the final step. Irrelevant
isotypeand concentration-matched primary mAbs were used as negative
control in all experiments.
Four biopsies from each subject were formalin-fixed and
paraffin-embedded in the same paraffin block. Sections were cut
in series at 4 mm and dewaxed. Heat-induced epitope retrieval
(HIER) was performed by boiling sections for 20 minutes in
TrisEDTA (pH = 9) in a water bath followed by cooling for 20 minutes
at RT. Immunoenzyme staining was performed with Ventana
Ultra View DAB Detection Kit (Ventana Medical Systems, Inc.,
Tucson, AZ). Sections were pre-blocked with H2O2 for 8 minutes.
Primary antibodies used were either monoclonal rabbit
antihuman CD3 (clone SP7, IgG, 1/100, Thermo Scientific, Fremont,
CA) or monoclonal mouse anti-human neutrophil elastase (clone
NP57, IgG1, 1/600, Dako) for 30 minutes in RT. When stained
with anti-neutrophil elastase (clone NP57) sections were not
pretreated with HIER. Antibody Diluent (S0809; Dako) replaced
the primary antibody as negative control. All sections were
counterstained with haematoxylin.
Evaluation of tissue staining results
Examination of immunofluorescence stainings was performed
with an epifluorescence microscope (Nikon Eclipse 80i, Nikon
Corporation, Tokyo, Japan). To determine the density of different
cell populations, all immunostained cells in lamina propria were
counted to a depth of ,0.5 mm from the basolateral side of the
surface epithelium. The area of lamina propria was estimated by
superimposing a grid (10610 lines; 0.24260.242 mm) parallel to
the muscularis mucosa. On average, 7 grids were examined for
every section. Combined fluorescent microscopy and differential
interference contrast microscopy (DIC) was used to visualize
eosinophils as previously described .
Immunoenzyme-stained formalin-fixed sections were examined
by light microscopy (Nikon Eclipse 50i, Nikon Corporation). To
determine the density of IELs, all intraepithelial CD3+ cells in the
upper half of the villi with satisfactory morphology were counted,
and the density was given as IELs per 100 epithelial cells (EPCs).
The density of neutrophils in lamina propria was determined by
superimposing a grid parallel to the muscularis mucosa as
All sections were examined at 400 X magnification by the same
investigator (A-C.R. Beitnes), blinded to patient identity and
Multicolor flow cytometry
Multiple biopsies (410) were collected in ice-chilled RPMI
medium and processed further in the laboratory within 30
minutes. Preparations of single cell suspensions were performed as
follows: EPCs and IELs were removed by incubation with 2 mM
EDTA in PBS twice for 30 minutes with continuous rotation at
37uC. Single-cell suspensions were obtained by digesting the
remaining material with 1 mg/mL Blend Collagenase (C-8051;
Sigma, St. Louis, MO) for 60 minutes with rotation at 37uC. The
cell suspension was then filtered through a 40 mm cell strainer and
washed with PBS. For analysis of blood cells, peripheral blood
mononuclear cells were isolated from acid-citrate-dextrose (ACD)
blood using standard protocols for density centrifugation with
Lymphoprep. Cells were transferred onto V-bottomed 96-well
plates, washed in PBS containing 0.5 mM
ethylenediaminetetraacetic acid (EDTA) and 3% foetal calf serum (FCS) and stained
with directly labeled antibodies on ice for 30 minutes. After
incubation, cells were briefly washed, resuspended in PBS with 3%
FCS and analyzed with a LSRII (BD Biosciences, Franklin Lakes,
NJ) instrument. To exclude dead cells, 0.2 mg/mL propidium
iodide (PI) was added to the samples immediately prior to analysis.
2246105 cells were analyzed in each sample.
The following antibodies and dilutions were used:
anti-CD45FITC, -PE, -APC and -Pacific Blue (all clone HI30, 1/20) and
anti-HLA-DR-eFluor450 (clone L243, 1/40) from eBioscience,
San Diego, CA; anti-CD11c-Alexa488 (clone 3.9, 1/20),
antiCD14-APC-Cy7 (clone HCD14, 1/20), anti-CCR2-Alexa 647
(clone TG5/CCR2, 1/20) and IgG2b-Alexa647 isotype control
(clone MPC11, 1/20) from Biolegend, San Diego, CA;
antiCD11c-PE (clone S-HCL-3, 1/15), anti-HLA-DR-PE-Cy7 (clone
L243, 1/100) from BD Biosciences and
anti-DC-SIGN/CD209PE (clone DCN46, 1/15) from BD Pharmingen, San Diego, CA.
T cell assay
Monocytes were isolated from peripheral blood mononuclear
cells using MACS CD14 MicroBeads (Miltenyi Biotec STED)
according to the manufacturers instructions. Recovered cells were
stained for CD14 and CD11c to assess purity. Monocytes (26104
cell) were transferred into 96-well plates and incubated overnight
with or without peptide antigen and with or without recombinant
IFN-c (100 U/ml; R&DSystems, Minneapolis, MN). Two gliadin
peptides harboring the HLA-DQ2.5-glia-a2 epitope were tested;
PQPELPYPQPQL (DQ2-a2) at 10 mM or
LQLQPFPQPELPYPQPELPYPQPELPYPQPQPF (33-mer) at 2 mM [13,14]. The
CD14+ monocytes were then washed and incubated with 50.000
cells of the gut derived T-cell clone TCC418.104.22.168 specific for the
HLA-DQ2.5-glia-a2 epitope. The gluten-reactive T-cell clone was
generated as previously described . T-cell proliferation was
evaluated after 72 hours by uptake of [3H]thymidine (1 mCi/well
(0.037 MBq/well); Hartmann Analytic, Braunshweig, Germany),
which was added to the wells 24 hours before harvesting with an
automated harvester (Mach III; TomTec, Hamden, CT).
Incorporated radioactivity was measured by liquid scintillation
counting (Wallac MicroBeta TriLux 1450; PerkinElmer,
Wellesley, MA). The assays were performed in triplicates.
Wilcoxon matched pairs test was used to compare the density of
different cell populations in the duodenal mucosa before and after
gluten challenge. GraphPad Prism 4 software (GraphPad
Software, La Jolla, CA) was used for statistical analysis.
CD14+CD11c+ DCs are selectively increased after a
threeday gluten challenge in CD patients
We found that the density of HLA-DQ+ APCs in lamina
propria was similar before and after gluten challenge in the
patients with treated CD (median 1755 cells/mm2, range 1398
2247 vs. median 1598, range 14172274, respectively). Notably,
however, the relative proportions of HLA-DQ+ APC subsets
defined by expression of the markers CD11c, CD163, CD14,
CD103 and CD1c were changed. By two-color
immunofluorescence staining we found an increase in the density of
CD14+CD11c+ (p = 0.001) and CD163+CD11c+ (p = 0.001)
DCs, and a decrease in the densities of CD103+ DCs (p = 0.03)
and CD163+CD11c2 macrophages (p = 0.03) (Figure 1). The
density of CD1c+ DCs was unchanged. We have recently shown
that CD14+CD11c+ DCs and CD163+CD11c+ DCs are mostly
overlapping populations both in the normal small intestinal
mucosa and in the active CD lesion . In the following
experiments we therefore used the co-expression of CD14 and
CD11c to identify this DC subset.
Possible mechanisms for CD14+CD11c+ DC accumulation
The phenotype of the accumulating DCs in the challenged
mucosa, being CD14+CD11c+CD163+, resembles that of CD14+
monocytes in peripheral blood . We therefore hypothesized
that the rapid accumulation of CD14+CD11c+ DCs was caused by
an increased recruitment of CD14+ monocytes from the
Figure 1. CD14+CD11c+ dendritic cells are selectively increased in duodenal mucosa of celiac disease patients after short-term
gluten challenge. Density of HLA-DQ+ antigen-presenting cell subsets in cryosections from celiac disease (CD) patients on gluten-free diet (GFD)
before and after a three-day gluten challenge. The density of CD163+CD11c2 macrophages is calculated by subtracting the number of
CD163+CD11c+ cells from the total number of CD163+HLA-DQ+ cells. Paired data are connected by lines. ns = not significant (A). Three-color
immunofluorescence staining for CD11c (green), CD14 (red) and cytokeratin (blue) in cryosection of duodenal mucosa from a CD patient on GFD
before and after a three-day gluten challenge. Original magnification X 400 (B).
circulation. Because the classical CD14+ monocytes, which
constitute the vast majority of all monocytes in the circulation,
can be identified by their expression of CCR2, we tested whether
CD14+CD11c+ DCs in the challenged mucosa expressed this
marker. In agreement with previous reports [17,18] we showed
that CCR2, as analyzed in two individuals, was highly expressed
on most peripheral blood CD14+ monocytes, whereas only a
fraction of myeloid DCs (CD142CD11c+) expressed CCR2
(Figure 2A). Consistent with this we found that the majority of
CD14+CD11c+ DCs in the intestinal mucosa also expressed
CCR2; both in healthy controls, treated CD and active CD
(Figures 2B and 3). In contrast, only a fraction of mucosal
CD142CD11c+ DCs and CD14+CD11c2 macrophages expressed
this marker (Figures 2B and 3). Together, these findings further
strengthen the notion that the CD14+CD11c+ DCs, which rapidly
accumulate in the duodenal mucosa in response to gluten
challenge, are recruited from circulating CD14+ monocytes.
CD14+CD11c+ DCs also expressed CD209 (dendritic cell-specific
intercellular adhesion molecule-3-grabbing non-integrin,
DCSIGN) both in healthy controls (n = 4; range 5676%), in an
untreated CD patient (77%), and a treated CD patient (73%)
(Figure 2C). As previously reported  we found that DC-SIGN
was highly expressed on macrophages, but virtually absent on
CD142CD11c+ DCs (Figure 2C).
Figure 2. Expression of CCR2 on HLA-DR+ leukocytes in blood and duodenal mucosa. Flow cytometric analysis of peripheral blood
mononuclear cells from treated celiac disease (CD) patient (A) and viable single cells of duodenal mucosa from treated CD patient showing the
expression of CCR2 (B) and DC-SIGN (C) on CD45+HLA-DR+ cells depending on the expression of CD11c and CD14. Dead cells were excluded by
adding 0.2 mg/mL propidium iodide immediately before acquisition. The data are representative for two independent experiments.
The increase of CD14+CD11c+ DCs precedes architectural
changes and increase of CD3+ IELs
The histopathology of the CD lesion is characterized by villous
blunting, crypt hyperplasia and increased numbers of IELs .
Next we wished to determine whether accumulation of
CD14+CD11c+ DCs preceded typical features of an established
celiac lesion. In the same biopsy material, Brottveit et al. recently
showed that in most patients there were no significant changes in
tissue architecture after a three-day gluten challenge . Only four
of twelve patients had changes according to Marsh classification.
As increased numbers of CD3+ IELs is one of the first signs of CD
[19,20], we counted CD3+ IELs in all samples. Importantly, no
difference in the number of CD3+ IELs was detected comparing
tissue obtained before and after challenge (Figure 4A). Thus, the
increase of CD14+CD11c+ DCs precedes the histopathological
hallmarks of an active CD lesion.
Increased recruitment of neutrophils and eosinophils are also
features of the CD lesion [21,22]. Of these immune cells only
neutrophils showed a modest but significant increase (p = 0.04)
after the short-term challenge (Figures 4B and 4C and Figure S1).
Increased neutrophil recruitment demonstrated that the exposure
to gluten triggered an inflammatory response.
CD14+CD11c+ monocytes from blood efficiently present
gluten to gluten-specific T-cell clones
We were not able to isolate CD14+CD11c+ DCs from the celiac
lesion to test their antigen-presenting capability in vitro. Instead, we
isolated CD14+CD11c+ monocytes from peripheral blood of two
HLA-DQ2+ individuals and tested the ability of these cells to
present gluten peptide antigen to an HLA-DQ2-restricted
glutenspecific T-cell clone. From both individuals, the purified
monocytes efficiently activated the T cells in an antigen
dosedependent manner. Preactivating the monocytes with IFN-c
further increased their antigen presenting capacity (Figure 5).
No changes of cell densities in gluten-sensitive controls
Next we wanted to examine whether the observed changes
might represent an innate response to gluten independently of CD.
To this end, we examined biopsy samples from gluten-sensitive
control subjects on GFD who were challenged with gluten as
described above. The gluten-sensitive controls were HLA-DQ2+
subjects without confirmed CD diagnosis, but with gluten-induced
symptoms that subjectively improved on a GFD. As recently
reported, no histopathological changes according to Marsh
classification or tetramer-positive CD4+ T cells in peripheral
blood were observed in this control group, although these patients
experienced symptoms during the three-day gluten challenge .
Interestingly, as opposed to CD patients, no significant changes in
the density of CD14+CD11c+ DCs, CD103+ DCs, neutrophils,
eosinophils or CD3+ IELs were found after challenge (Figure 6).
This finding indicates that the rapid recruitment of
Figure 5. Peripheral blood CD14+ monocytes efficiently present gluten to gluten-specific T cell clones. Purity of CD14+CD11c+
monocytes isolated from peripheral blood mononuclear cells of two individuals are shown (A and B, upper panels). The monocytes were incubated
with medium or two different gluten peptides 6100 U/ml IFN-c for 24 hours, washed and incubated with a T-cell clone for 72 hours. The proliferative
T-cell response (measured by thymindine-incorporation) is shown (A and B, lower panels). Experiments with IFN-c are indicated (hatched columns).
We have studied changes in duodenal mucosal APC populations
after a three-day gluten challenge in CD patients in remission and
in gluten-sensitive control subjects. In the CD patients we found
alterations in the APC populations in response to the challenge
that preceded morphological changes in most participants. These
changes resulted in a relative composition of APCs similar to that
found in untreated CD  and were not observed in the
glutensensitive controls. The findings suggest that the changes in the
duodenal APC composition occur rapidly to gluten exposure and
represent an early and integrated part of the immune reaction
leading to CD.
The most pronounced difference was observed for the
population of CD14+CD11c+ DCs. The density of these cells
was increased almost three-fold in the celiac duodenal mucosa
after the three-days gluten challenge, similar to that found in
established celiac lesions . Most notably, the increase preceded
the typical architectural changes as well as an increase of IELs and
eosinophils, which suggests that these cells may be important for
disease development. There is a striking phenotypic resemblance
between the DC subset that accumulated in the challenged
mucosa and the classical monocytes in peripheral blood. Both cell
populations express CD14, CD11c, CD163 and CCR2, which is a
unique combination of markers in both blood and tissue. Although
not formally demonstrated it is therefore tempting to speculate
that the CD14+CD11c+ DCs accumulating in the tissue are
derived from circulating CD14+ monocytes . CD14+CD11c2
macrophages had lower expression of CCR2, suggesting that
resident macrophages originating from monocytes might
downregulate CCR2 .
An increasing body of evidence suggests that monocytes have
the capacity to differentiate into efficient DCs in tissues. It was
recently reported that in mice the expression of DC-SIGN/
CD209 distinguishes monocyte-derived DCs (Mo-DCs) from
classical DCs both in cell suspension and lymph nodes.
Furthermore, these Mo-DCs (being CD14+CD11c+) were
demonstrated to have strong antigen-presenting activity . Most
notably, we found that most CD14+CD11c+ DCs in the duodenal
mucosa also express DC-SIGN/CD209, which is a putative
marker of fully differentiated Mo-DCs. It is therefore conceivable
that the recruited CD14+CCR2+ monocytes rapidly differentiate
into efficient HLA-DQ+ APCs that activate gluten-reactive T cells
residing in the intestinal mucosa. In agreement with this notion,
we demonstrated that CD14+CD11c+ monocytes isolated from
peripheral blood presented gluten peptides efficiently to
glutenspecific T cells in a HLA-DQ-restricted manner. This is in line
with our previous data showing that CD11c+ cells from duodenal
biopsies are capable of presenting antigen to gluten specific T cells
in vitro .
The gluten-sensitive subjects served as controls as they
experienced gluten related symptoms, which we consider relevant,
and as we were unable to recruit a control group of completely
healthy subjects who adhered to a strict GFD. The mechanisms
leading to symptoms in gluten intolerant patients are poorly
understood. While integrated innate and adaptive immune
responses appear important in the pathogenesis of CD , it is
speculated whether symptoms of gluten intolerant patients may
result from an unaccompanied innate immune response to gluten
. However, conflicting results are reported on the effect of
Figure 6. Density of various leukocyte subsets remains unchanged in duodenal mucosa of gluten-sensitive control subjects after
short-term gluten challenge. Density of CD14+CD11c+ dendritic cells (DCs) per mm2 (A) and CD103+CD11c+ DCs per mm2 (B) in the lamina propria
(LP); CD3+ intraepithelial lymphocytes (IELs) per 100 epithelial cells (C); neutrophils per mm2 (D) and eosinophils per mm2 (E) in the LP in sections of
duodenal mucosa from gluten-sensitive control subjects on gluten-free diet before and after a three-day gluten challenge. Paired data are connected
by lines. ns = not significant.
gluten on the innate immune system. It has been reported that in vitro
challenge of CD14+ human monocytes with digested gliadin causes
maturation of the monocytes into DCs regardless of genetic
predisposition or presence of CD in the cell donors [27,28]. This
gives credence to the above notion, but is at variance with the
observation that stimulation of human duodenal biopsies with
digested gliadin only gives activation of the innate immune system
in CD patients and not in healthy controls . Neutrophils and
APCs both belong to the innate immune system, and in our study we
observed early changes in these cell populations in the CD patients. In
the established celiac lesion, the density of neutrophils and
CD14+CD11c+ APCs are reported to increase [5,21,30]. Therefore,
it was of interest to look at these cell populations in particular in
gluten-sensitive control subjects before and after a short-term gluten
challenge. The fact that the density of either APCs or neutrophils
changed significantly in gluten-sensitive controls after challenge,
demonstrates that the changes in these cell populations are restricted
to CD. Moreover, the findings suggest that the symptoms reported in
these patients upon the three-day gluten challenge, likely do not relate
to innate immune activation by gluten in APCs or neutrophils.
The cues that lead to changes in the composition of the APC
subpopulations in CD could possibly involve activation of T cells.
Experiments in vitro have demonstrated that CD3+ T cells from
duodenal biopsies of CD patients challenged with peptic-tryptic
gluten digest or a gliadin fragment were activated when harvested
after 24 hours as shown by upregulation of CD25 [29,31]. Thus, it
is to be expected that T cell activation takes place within the time
frame of three days. Moreover, IFN-c is shown to be produced
both in biopsies and CD4+ T cells from CD patients upon
24 hours in vitro gluten challenge [32,33]. Conceivably, in treated
CD patients there might be interaction between primed T cells
and APCs taking place at an early stage after gluten challenge.
The observation that gluten exposure does not lead to changes in
neutrophils and APC subpopulations in gluten-sensitive control
subjects supports the model that activation of gluten specific T cells
is implicated in the innate immune response in CD.
Taken together, we have found that CD14+CD11c+ DCs
rapidly and selectively increase in the gluten challenged duodenal
mucosa of treated CD patients. This did not occur in
glutensensitive control subjects, making the gluten-induced recruitment
of CD14+CD11c+ DCs specific for CD. Accumulation of this DC
subset prior to induction of architectural changes and increase in
IELs suggests that they are directly involved in the
immunopathology of CD.
Figure S1 Immunostaining of neutrophils in duodenal
mucosa. Immunoenzyme staining for neutrophil elastase to
visualize neutrophils (arrows) on formalin-fixed and
paraffinembedded sections from duodenal mucosa of normal individuals
The authors express gratitude to the patients for donating duodenal
biopsies and the staff at the Endoscopy Unit, Department of Medicine,
Oslo University Hospital Rikshospitalet, for assistance in collecting
biopsies. The authors thank Aaste Aursj, Hogne Red Nilsen, Linda
Manley, Kjersti Thorvaldsen Hagen, Jorunn Elisabet Bratlie and Marie
Kongshaug Johannesen for excellent technical training and assistance.
Conceived and designed the experiments: ACRB MR MB KEAL FLJ
LMS. Performed the experiments: ACRB MR. Analyzed the data: ACRB
MR FLJ LMS. Contributed reagents/materials/analysis tools: MB KEAL
FLJ LMS. Wrote the paper: ACRB FLJ LMS.
1. Sollid LM ( 2002 ) Coeliac disease: dissecting a complex inflammatory disorder . Nat Rev Immunol 2 : 647 - 655 .
2. Jabri B , Sollid LM ( 2009 ) Tissue-mediated control of immunopathology in coeliac disease . Nat Rev Immunol 9 : 858 - 870 .
3. Walker MM , Murray JA ( 2011 ) An update in the diagnosis of coeliac disease . Histopathology 59 : 166 - 179 .
4. Raki M , Tollefsen S , Molberg O , Lundin KE , Sollid LM , et al. ( 2006 ) A unique dendritic cell subset accumulates in the celiac lesion and efficiently activates gluten-reactive T cells . Gastroenterology 131 : 428 - 438 .
5. Beitnes AC , Raki M , Lundin KE , Jahnsen J , Sollid LM , et al. ( 2011 ) Density of CD163(+) CD11c(+) dendritic cells increases and CD103(+) dendritic cells decreases in the coeliac lesion . Scand J Immunol 74 : 186 - 194 .
6. Brottveit M , Raki M , Bergseng E , Fallang LE , Simonsen B , et al. ( 2011 ) Assessing possible celiac disease by an HLA-DQ2-gliadin Tetramer Test . Am J Gastroenterol 106 : 1318 - 1324 .
7. Oberhuber G , Granditsch G , Vogelsang H ( 1999 ) The histopathology of coeliac disease: time for a standardized report scheme for pathologists . Eur J Gastroenterol Hepatol 11 : 1185 - 1194 .
8. Spits H , Borst J , Giphart M , Coligan J , Terhorst C , et al. ( 1984 ) HLA-DC antigens can serve as recognition elements for human cytotoxic T lymphocytes . Eur J Immunol 14 : 299 - 304 .
9. Micklem KJ , Dong Y , Willis A , Pulford KA , Visser L , et al. ( 1991 ) HML-1 antigen on mucosa-associated T cells, activated cells, and hairy leukemic cells is a new integrin containing the beta 7 subunit . Am J Pathol 139 : 1297 - 1301 .
10. Wang JE , Warris A , Ellingsen EA , Jorgensen PF , Flo TH , et al. ( 2001 ) Involvement of CD14 and toll-like receptors in activation of human monocytes by Aspergillus fumigatus hyphae . Infect Immun 69 : 2402 - 2406 .
11. Huitfeldt HS , Brandtzaeg P ( 1985 ) Various keratin antibodies produce immunohistochemical staining of human myocardium and myometrium . Histochemistry 83 : 381 - 389 .
12. Jahnsen FL , Haraldsen G , Rugtveit J , Halstensen TS , Brandtzaeg P ( 1994 ) Differential interference contrast microscopy combined with immunofluorescence: a new method to phenotype eosinophils in situ . J Immunol Methods 173 : 77 - 91 .
13. Arentz-Hansen H , Korner R , Molberg O , Quarsten H , Vader W , et al. ( 2000 ) The intestinal T cell response to alpha-gliadin in adult celiac disease is focused on a single deamidated glutamine targeted by tissue transglutaminase . J Exp Med 191 : 603 - 612 .
14. Shan L , Molberg O , Parrot I , Hausch F , Filiz F , et al. ( 2002 ) Structural basis for gluten intolerance in celiac sprue . Science 297 : 2275 - 2279 .
15. Molberg O , McAdam SN , Lundin KE , Sollid LM ( 2000 ) Studies of gliadinspecific T-cells in celiac disease . Methods Mol Med 41 : 105 - 124 .
16. Wong KL , Tai JJ , Wong WC , Han H , Sem X , et al. ( 2011 ) Gene expression profiling reveals the defining features of the classical, intermediate, and nonclassical human monocyte subsets . Blood 118 : e16 - e31 .
17. Tallone T , Turconi G , Soldati G , Pedrazzini G , Moccetti T , et al. ( 2011 ) Heterogeneity of human monocytes: an optimized four-color flow cytometry protocol for analysis of monocyte subsets . J Cardiovasc Transl Res 4 : 211 - 219 .
18. Shantsila E , Wrigley B , Tapp L , Apostolakis S , Montoro-Garcia S , et al. ( 2011 ) Immunophenotypic characterisation of human monocyte subsets: Possible implications for cardiovascular disease pathophysiology . J Thromb Haemost 9 : 1056 - 1066 .
19. Leigh RJ , Marsh MN , Crowe P , Kelly C , Garner V , et al. ( 1985 ) Studies of intestinal lymphoid tissue . IX. Dose-dependent, gluten-induced lymphoid infiltration of coeliac jejunal epithelium . Scand J Gastroenterol 20 : 715 - 719 .
20. Marsh MN ( 1992 ) Gluten, major histocompatibility complex, and the small intestine. A molecular and immunobiologic approach to the spectrum of gluten sensitivity ('celiac sprue') . Gastroenterology 102 : 330 - 354 .
21. Dhesi I , Marsh MN , Kelly C , Crowe P ( 1984 ) Morphometric analysis of small intestinal mucosa . II. Determination of lamina propria volumes; plasma cell and neutrophil populations within control and coeliac disease mucosae . Virchows Arch A Pathol Anat Histopathol 403 : 173 - 180 .
22. Marsh MN , Hinde J ( 1985 ) Inflammatory component of celiac sprue mucosa . I. Mast cells, basophils, and eosinophils. Gastroenterology 89 : 92 - 101 .
23. Schwartzkopff F , Petersen F , Grimm TA , Brandt E ( 2010 ) CXC chemokine ligand 4 (CXCL4) down-regulates CC chemokine receptor expression on human monocytes . Innate Immun.
24. Cheong C , Matos I , Choi JH , Dandamudi DB , Shrestha E , et al. ( 2010 ) Microbial stimulation fully differentiates monocytes to DC-SIGN/CD209(+) dendritic cells for immune T cell areas . Cell 143 : 416 - 429 .
25. Schuppan D , Junker Y , Barisani D ( 2009 ) Celiac disease: from pathogenesis to novel therapies . Gastroenterology 137 : 1912 - 1933 .
26. Troncone R , Jabri B ( 2011 ) Coeliac disease and gluten sensitivity . J Intern Med 269 : 582 - 590 .
27. Rakhimova M , Esslinger B , Schulze-Krebs A , Hahn EG , Schuppan D , et al. ( 2009 ) In vitro differentiation of human monocytes into dendritic cells by peptictryptic digest of gliadin is independent of genetic predisposition and the presence of celiac disease . J Clin Immunol 29 : 29 - 37 .
28. Palova-Jelinkova L , Rozkova D , Pecharova B , Bartova J , Sediva A , et al. ( 2005 ) Gliadin fragments induce phenotypic and functional maturation of human dendritic cells . J Immunol 175 : 7038 - 7045 .
29. Maiuri L , Ciacci C , Ricciardelli I , Vacca L , Raia V , et al. ( 2003 ) Association between innate response to gliadin and activation of pathogenic T cells in coeliac disease . Lancet 362 : 30 - 37 .
30. Diosdado B , van BH , Strengman E , Franke L , van OE , et al. ( 2007 ) Neutrophil recruitment and barrier impairment in celiac disease: a genomic study . Clin Gastroenterol Hepatol 5 : 574 - 581 .
31. Halstensen TS , Scott H , Fausa O , Brandtzaeg P ( 1993 ) Gluten stimulation of coeliac mucosa in vitro induces activation (CD25) of lamina propria CD4+ T cells and macrophages but no crypt-cell hyperplasia . Scand J Immunol 38 : 581 - 590 .
32. Raki M , Molberg O , Tollefsen S , Lundin KE , Sollid LM ( 2005 ) The effects of atorvastatin on gluten-induced intestinal T cell responses in coeliac disease . Clin Exp Immunol 142 : 333 - 340 .
33. Bodd M , Raki M , Tollefsen S , Fallang LE , Bergseng E , et al. ( 2010 ) HLA-DQ2- restricted gluten- reactive T cells produce IL-21 but not IL-17 or IL-22. Mucosal Immunol 3 : 594 - 601 .