P38 MAPK signaling mediates retinoic 
acid‐induced CD103 expression in human 
dendritic cells
Authors: Mandi M. Roe, Marziah Hashimi, Steve 
Swain, Krista M. Woo, and Diane Bimczok 
This is the peer reviewed version of the following article: cited below, which has been published 
in final form at doi: 10.1111/imm.13246. This article may be used for non-commercial purposes 
in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions. This article 
may not be enhanced, enriched or otherwise transformed into a derivative work, without express 
permission from Wiley or by statutory rights under applicable legislation. Copyright notices must 
not be removed, obscured or modified. The article must be linked to Wiley’s version of record on 
Wiley Online Library and any embedding, framing or otherwise making available the article or 
pages thereof by third parties from platforms, services and websites other than Wiley Online 
Library must be prohibited.
Roe, Mandi M., Marziah Hashimi, Steve Swain, Krista M. Woo, and Diane Bimczok. “P38 MAPK 
Signaling Mediates Retinoic Acid‐induced CD103 Expression in Human Dendritic Cells.” 
Immunology 161, no. 3 (September 14, 2020): 230–244. doi:10.1111/imm.13246.
Made available through Montana State University’s ScholarWorks 
scholarworks.montana.edu 
DR. DIANE  BIMCZOK (Orcid ID : 0000-0002-8817-7243)
Article type : Original Article
P38 MAPK Signaling Mediates Retinoic Acid-Induced CD103 Expression in Human 
Dendritic Cells
Short title: Retinoic acid signaling in human DCs
Mandi M. Roe, Marziah Hashimi, Steve Swain, Krista M. Woo, and Diane Bimczok*
Department of Microbiology and Immunology, Montana State University, Bozeman, MT, 
USA
*Corresponding author: Dr. Diane Bimczok
Email: diane.bimczok@montana.edu
Phone: +1-406-994-4928
Author Contributions: M.M.R.: developed the project, designed and performed 
experiments, analyzed data, critically interpreted the data, wrote and revised the 
manuscript; M.H.: performed experiments, analyzed data, revised the manuscript; S.S.: 
This is the author manuscript accepted for publication and has undergone full peer review but 
has not been through the copyediting, typesetting, pagination and proofreading process, 
which may lead to differences between this version and the Version of Record. Please cite this 
article as doi: 10.1111/IMM.13246
This article is protected by copyright. All rights reserved
planned experiments, analyzed data, revised the manuscript; K.M.W.: performed 
experiments; D.B.: developed the project, designed the experiments, analyzed the data, 
critically interpreted the data, wrote and revised the manuscript, provided funding for the 
project.
Keywords: Dendritic cells, retinoic acid, CD103, p38 MAPK
Abbreviations:
RA – retinoic acid
DC – dendritic cell
MoDC – monocyte-derived dendritic cell
Integrin E – CD103
RAR – retinoic acid receptor
RARE – retinoic acid response elements
ITGAE – CD103 gene name
ITGB7 – 7 gene name
TGFRII – TGF-receptor II
Summary
Retinoic acid (RA) is an active derivative of vitamin A and a key regulator of 
immune cell function. In dendritic cells (DCs), RA drives the expression of CD103 
(integrin E), a functionally relevant DC subset marker. In this study, we analyzed the cell
type specificity and the molecular mechanisms involved in RA-induced CD103 expression. 
We show that RA treatment caused a significant upregulation of CD103 in differentiated 
monocyte-derived DCs and blood DCs, but not in differentiated monocyte-derived 
This article is protected by copyright. All rights reserved
macrophages or T cells. DC treatment with an RARα agonist lead to a similar increase in
CD103 expression as RA treatment, while RARA gene silencing with siRNA blocked RA-
induced upregulation of CD103, pointing to a major role of RARα in the regulation of
CD103 expression. To elucidate RA-induced signaling downstream of RARα, we used
Western blot analysis of RA-treated DCs and showed a significant increase of p38 MAPK 
phosphorylation. In addition, DCs cultured with RA and a p38 MAPK inhibitor had a 
significantly reduced expression of CD103 compared with DCs cultured with RA only, 
indicating that p38 MAPK is involved in CD103 regulation. In summary, these findings 
suggest that the RA-induced expression of CD103 is specific to DCs, is mediated primarily 
through RAR and involves p38 MAPK signaling.
Introduction
Retinoic acid (RA) is the major active derivative of vitamin A, an essential dietary 
micronutrient in humans and other mammals. RA has long been recognized as a key factor 
in embryonic tissue development and the maintenance of ocular function and plays 
important roles in mucosal immunity (1-3). Specifically, RA is involved in regulating 
lymphocyte homing to mucosal sites (4-6), induction of regulatory T cells (7), and mucosal 
dendritic cell (DC) function and development (4, 8). The relevance of Vitamin A and its 
active mediator RA for gastrointestinal immune function is illustrated by the fact that 
vitamin A deficiency leads to an increased risk for diarrheal infections and inflammatory 
bowel disease (9-11). 
Importantly, a significant number of animal studies have revealed an association 
between DC production of RA, the DC response to RA, and DC expression of CD103, the 
α chain of E7 integrin (7, 8, 12-14). CD103 is a functionally important subset marker for
gastrointestinal DCs in mice and is also expressed on mucosal DC subsets in humans (7, 
15, 16). Analyses from our laboratory and others have demonstrated that RA drives the 
expression of CD103 (E integrin) as well as the corresponding  chain 7 integrin on
human DCs (13, 17, 18). Several studies in mice suggest that signaling through retinoic 
acid receptor alpha (RAR), one of the three retinoic acid receptors (RAR), is the major
This article is protected by copyright. All rights reserved
t
pathway for RA-mediated regulation of DC development and function (4, 19); however, 
whether the same is true for human DCs is unclear. 
RA exhibits pleotropic effects through both genomic and non-genomic signaling 
pathways (20-23). Canonical genomic RA signaling involves direct modulation of gene 
expression through binding of RARs that function as transcription factors to RA response 
elements (RARE) on the DNA (24-26). Alternatively, RA can also elicit non-genomic 
effects through interactions with cytoplasmic signaling cascades such as p38 MAPK and 
ERK1 signaling (21, 23, 27). Our previous analyses showed that RA signaling leading to 
CD103 expression in human DCs intersects with TGF-β signaling, since a small molecule
inhibitor of the TGF- receptor prevented RA-induced CD103 upregulation (13). In
activated CD8 T cells, TGF- leads to a strong induction of CD103 expression (28, 29).
However, TGF- alone does not drive CD103 expression in DCs, as we and others have
shown ointing to complex interactions between RA signaling and TGF-
signaling pathways. Overall, these observations led us to investigate the non-canonical 
signaling pathways downstream of RA that may be responsible for the regulation of CD103 
expression.
We here investigated the cell type specificity of RA-induced CD103 expression and 
the molecular mechanisms that drive RA-dependent expression of CD103 in human DCs. 
Our experiments showed that RA increased expression of CD103 in MoDCs and blood 
DCs and to a lesser extent in monocytes, but not in differentiated macrophages or T cells. 
Moreover, induction of CD103 was dependent on signaling through RAR and p38
MAPK, but not SMAD2/3. We also showed that inhibition of NFAT diminished the 
expression of CD103 in DCs in the presence of RA, suggesting that NFAT may be a 
downstream target of p38 MAPK in RA-dependent signaling.  
Materials and Methods:
Human donors: 
Whole blood was obtained from a pool of approximately 20 healthy male and 
female volunteers aged 21 - 60 with informed consent. For each experiment, 3 – 5 donors 
This article is protected by copyright. All rights reserved
were randomly selected from this pool. Collection of human blood samples was approved 
by Montana State University’s Institutional Review Board (IRB), protocol #DB082817.
Cell isolation and culture: 
CD14+ monocytes were isolated from heparinized whole blood by centrifugation 
and MACS sorting, as previously shown (13). Monocytes were cultured in X-vivo media 
(Lonza, Basel, Switzerland) supplemented with 100 U/l penicillin, 100 g/l streptomycin,
50 g/ml gentamycin, 5mM HEPES, and 2 mM L. glutamine (all Hyclone, Logan, UT,
USA). To induce differentiation of a monocyte-derived macrophage phenotype, 25 ng/ml 
recombinant human (rh)M-CSF (PeproTech, Rocky Hill, NJ, USA) was added to the 
culture medium. To induce differentiation of monocyte-derived DCs (MoDCs), 25 ng/ml 
rhGM-CSF and 7 ng/ml rhIL-4 (R&D Systems, Minneapolis, MN, USA) were added to the 
culture medium. Aliquots of all monocyte samples were tested for TNF secretion after 24
hours using an ELISA (R&D Systems, Minneapolis, MN), and monocytes with 
spontaneous TNF secretion >100 pg/mL were considered pre-activated and thus excluded
from the analyses, as described previously (13, 14). Unretained, monocyte-depleted cells 
from CD14 MACS sorting were used for lymphocyte experiments and were also cultured in 
X-vivo media. To induce T cell proliferation, Human T-Activator CD3/CD28 Dynabeads®
(Thermofisher Scientific, Waltham, MA, USA) were added to the lymphocyte cultures at 
0.5 µL/mL, and proliferation was determined by dye dilution assay using CellTraceViolet 
(Thermofisher Scientific). Human blood DCs were isolated from PBMCs by MACS using 
the Blood Dendritic Cell Isolation Kit II (Miltenyi Biotec, Bergisch Gladbach, Germany). 
Reagents and antibodies: 
RA was obtained from Sigma-Aldrich (St. Louis, MO, USA) and was used at a 
physiological concentration of 100 nM. All RA-treated cells were handled under red light 
to prevent RA degradation. The following RA agonists and antagonists were applied to 
This article is protected by copyright. All rights reserved
i
MoDC cultures during the first 3 days of culture: RAR agonist AM80 (1-10 M), RAR
agonist CD2314 (1-10 M), RAR agonist CD437 (1-10 M), RAR antagonist
BMS195614 1-50 nM; all from Tocris Bioscience, Bristol, UK) and RA antagonist Ro41-
5253 (60 nM, Sigma-Aldrich). Intracellular signaling pathways were targeted by applying 
the following reagents to mature MoDC cultures for 1 hour at 37ºC before the addition of 
RA: The p38 MAPK inhibitor SB202190 (10 M, Sigma-Aldrich) and an NFAT peptide
inhibit M, Tocris Bioscience). The following anti-human monoclonal antibodies
were used for flow cytometry: CD103 (integrin αE), integrin 7 (both eBioscience, Inc.,
San Diego, CA, USA), CD3, phopho-SMAD2/3, CD13, CD14, CD80 (all BD Biosciences, 
San Jose, CA, USA), CD4 (Tonbo Biosciences, San Diego, CA, USA), and CD1c and 
CD141 (both BioLegend, San Diego, CA). A fixable live/dead stain was applied to all flow 
cytometry samples (ThermoFisher Scientific, Waltham, MA, USA). The following 
antibodies were used for Western blot analysis: p38 MAPK, pThr180/182 p38 MAPK, and 
GAPDH (all Cell Signaling, Danvers, MA, USA).
Flow cytometry:
Cells were stained with the antibodies listed above. Isotype controls for each 
antibody were used to control for non-specific binding. We used a live/dead yellow fixable 
stain to exclude non-viable cells from our analyses. Surface antibody staining was 
performed as described previously (13). Following antibody staining, MoDCs were fixed 
with Cytofix (BD Biosciences) and analyzed on an LSR II or an LSR Fortessa Flow 
Cytometer (BD Biosciences). FACS data were analyzed using FlowJo X software (Tree 
Star, Ashland, OR, USA). MoDCs were gated based upon size, single cells, and live cells. 
For intracellular detection of phosphorylated SMAD2/3, MoDCs were cultured for 3 days, 
harvested, and allowed to rest for 2 hours prior to stimulation with RA, 5 ng TGF-β1 (R&D
Systems), or TGF-β1 and 50 M TGF-βRII inhibitor, SB431542 (Tocris Bioscience) for 30
min. Cells then were incubated in pre-warmed Cytofix (BD Biosciences) at 37ºC for 10 
This article is protected by copyright. All rights reserved
i
min. Cells were washed and resuspended in Perm Buffer III (Fisher Scientific) for 30 min 
at 4ºC and then were processed for staining with antibodies.
RAR siRNA knock down:
siRNA specific to RAR was used to inhibit RAR expression in MoDCs. MoDCs
were differentiated for three days before treatment. MoDCs then were resuspended in BTX 
electroporation solution (BTX, Holliston, MA, USA) and 5 g RAR siRNA or control
scramble siRNA (Santa Cruz Biotechnology, Dallas, TX, USA) and were electroporated 
using program U002 for eukaryotic cells on NucleofectorTM 2b device (Lonza). MoDCs 
were immediately added to pre-warmed media and incubated for 24 hours before the 
addition of RA. Following 24 hours of culture with RA, MoDCs were harvested for 
quantitative reverse transcriptase PCR (qRT-PCR) analysis of RAR, CD103, and integrin
7 expression.
Western blot:
MoDCs cultured in the presence of RA for 3 days were harvested for Western blot 
into RIPA lysis buffer for 30 min at 4°C, then centrifuged at 10,000 g for 15 min. 
Supernatant was recovered and mixed with Laemmli sample buffer (Bio-Rad Laboratories, 
Hercules, CA, USA). Protein was quantified with a BCA assay following the 
manufacturer’s protocol (Thermo Fisher Scientific). Equal quantities of protein, 1-3 g per
lane, were run on a 10% SDS-PAGE gel for 1 hour at 140 V. Protein was transferred to a 
PVDF membrane (Bio-Rad Laboratories) for 40 minutes at 100 V. Membranes were 
blocked with TBST, 5% BSA, sodium fluoride, proteinase inhibitors, and phosphatase 
inhibitors (all Thermo Fisher Scientific). Membranes were incubated overnight with the 
This article is protected by copyright. All rights reserved
primary antibody, washed, and then incubated with the secondary antibody conjugated with 
horseradish peroxidase for 1 hour at room temperature. Membranes were developed with 
SuperSignalTM West Pico PLUS (Thermo Fisher Scientific) per manufacturer’s 
specifications and then were imaged on a FluorChemR system (Protein Simple, San Jose, 
CA, USA). Band intensity was quantified using ImageJ 1.52p software (30).
Gene expression analysis by quantitative RT-PCR: 
RNA extraction from MoDCs was performed with a Direct-zol kit (Zymo Research, 
Irvine, CA, USA) per manufacturer’s specifications. RNA was quantified using a 
Nanodrop1000 (Thermo Fisher Scientific). An iScriptTM cDNA synthesis kit (Bio-Rad 
Laboratories) was used to generate cDNA. TaqMan Mastermix and primer/probes (Thermo 
Fisher Scientific) for each gene of interest and GAPDH housekeeping gene were used for 
qRT-PCR, as previously described (13). Samples were amplified on a Lightcycler®96 
(Roche Holding AG, Basel, Switzerland), using the following protocol: 1 cycle for UNG 
activation at 50ºC for 2 minutes. 1 cycle for hot start at 95ºC for 10 minutes. 40 cycles of 
amplification as follows: 95ºC for 15 seconds, 60ºC for 1 minute. Fluorescence was 
measured after each of these cycles. Data analysis was performed using the Pfaffl method 
(31).
Statistics:
Data were analyzed using GraphPad Prism 8.3.1 (GraphPad Software, San Diego, 
CA, USA). Results are presented as individual data points and/or mean ± standard 
deviation (SD). Differences between values were analyzed for statistical significance by 
Student’s t test, the non-parametric Kruskall-Wallis test, one-way or two-way ANOVA 
This article is protected by copyright. All rights reserved
with appropriate multiple comparisons tests. Differences were considered significant at 
P<0.05.
Results:
RA induces CD103 expression in monocytes and monocyte-derived dendritic cells, but not 
in differentiated monocyte-derived macrophages. 
In a recent study, we demonstrated that human monocyte-derived DCs (MoDCs) 
that were differentiated in the presence of RA for three days significantly upregulated 
CD103 and β7 integrin expression (13). To determine whether the ability to upregulate
CD103 was specific to DCs and required early, long term RA exposure, we here performed 
parallel experiments with monocytes, monocyte-derived macrophages (MDMs) and 
MoDCs with varying RA exposure protocols. MACS-purified CD14+ monocytes were 
treated with RA alone or in combination with either GM-CSF and IL-4 to induce 
differentiation to MoDCs or in combination with M-CSF to induce differentiation to 
MDMs. Interestingly, a one-day exposure of freshly isolated blood monocytes to RA was 
sufficient to increase the percentage of CD103-expressing cells in monocytes and to a 
greater extent in monocytes cultured in the presence of DC or macrophage-polarizing 
cytokines (P<0.01) (Fig. 1A-C). We next differentiated monocytes into either MoDCs or 
MDMs for 3 days prior to adding RA for another 24 h (Fig. 1D).  Differentiation was 
confirmed based on significantly decreased CD14 expression in the MoDCs compared to 
monocytes and MDMs (P=0.008, Suppl. Fig. 1A, B), and significant upregulation of CD13 
in both MoDCs and MDMs (P=0.009, Suppl. Fig. 1C) (32). Interestingly, RA treatment of 
all three cell types led to a significantly increased expression of CD14 (P=0.004, Suppl. 
Fig. 1B). However, in contrast to our previous study, where MoDCs that were 
differentiated in the presence of RA showed decreased expression of HLA-DR, CD83 and 
CD86 (13), a one-day exposure of differentiated MoDCs to RA failed to downregulate 
HLA-DR (Suppl. Fig. 1D) or co-stimulatory molecules (not shown). Importantly, as shown 
in Fig. 1E,F, a 24 h exposure to RA induced a significant increase in CD103 expression in 
This article is protected by copyright. All rights reserved
differentiated MoDCs, but not in MDMs. These observations indicate that RA-dependent 
induction of CD103 is cell type specific, with DCs being particularly responsive to RA-
treatment. Importantly, the data shown here together with the data from our previous study 
(13) indicate that the ability of RA to induce CD103 expression is highly robust regarding
timing and length of RA exposure.
RA does not induce CD103 expression in CD4+ or CD8+ T cells 
Previous reports have shown that TGF- can induce CD103 expression on T cells
(28, 33-35), but neither TGF- nor TGF- induced CD103 expression by human DCs in
our previous report (13), again pointing to cell type-specific regulatory mechanisms for 
CD103 induction. To determine whether RA would induce the expression of CD103 
expression in T cells, we treated lymphocytes with RA. Following 24 hours of culture, the 
cells were harvested and stained for T cell markers and CD103 (Fig. 2A). Gating strategies 
used to identify CD4+ and CD8+ T cell populations are shown in Fig. 2B. Interestingly, RA 
did not significantly increase the expression of CD103 on CD4+ or CD8+ T cells (Fig. 2C, 
D). A small increase of CD103 expression was seen in CD8+ T cells treated with TGF-
or TGF- (Fig. 2D), consistent with previous reports (28, 33, 34), but TGF- had no
apparent effect on CD103 expression by CD4+ T cells (Fig. 2C). Similarly, no CD103 
expression was detected with any of the treatments in CD14-depleted CD3− cells, which
contain a large proportion of B cells (data not shown). 
We next analyzed whether T cells respond to RA after polyclonal stimulation with 
anti-CD3/CD28 beads (Fig. 2E). The presence of RA in addition to the anti-CD3/CD28 
significantly increased proliferation of CD8+ T cells (P≤0.01) and slightly increased
proliferation of CD4+ T cells (Fig. 2F, G). However, CD103 expression in both CD4+ and 
CD8+ T cells remained unchanged, regardless of experimental condition (Fig. 2H, I), 
although CD103 expression overall was higher in cells that had divided.  In summary, these 
data suggest that the ability of RA to induce CD103 expression is specific to the myeloid 
cell compartment, particularly DCs.
This article is protected by copyright. All rights reserved
RA-induced expression of CD103 in human MoDCs is mediated primarily by RARα.
RA signaling is mediated through three different retinoic acid receptor (RAR) 
subtypes with different expression patterns and target genes, RAR RAR, or RAR (36).
In order to elucidate the RA signaling pathway involved in RA-induced CD103 expression, 
we used agonists specific for the different RARs, which we added to MoDC cultures for 3 
days during MoDC differentiation (Fig. 3A). Treatment of MoDCs with 1 and 10 M
AM80, an RAR-specific agonist, induced a similar increase in the percentage of CD103+
cells as treatment with 100 nM of RA (Fig. 3B, D). In contrast, the RAR agonist CD2314
did not increase the expression of CD103 in MoDCs, while treatment with the RAR
agonist CD437 caused a significant increase at 10 M, but not at 1 M (Fig. 3B, D).
Similar patterns were observed when geometric mean fluorescence intensity was analyzed 
(Suppl. Table 1). CD103 (integrin E) generally forms heterodimers with integrin 7, and
we have previously shown that CD103 and integrin 7 had similar expression patterns
when MoDCs were cultured with RA (13). However, none of the RAR agonists caused a 
significant increase in integrin 7 expression, although the overall 7 expression pattern in
response to the agonist treatment was similar to that of CD103 (Fig. 3C and Suppl. Table 
2).
Having demonstrated that signaling through RAR is most efficient at inducing
CD103 expression in DCs, we next used siRNA knockdown of RARA to inhibit RARα
signaling. For knockdown experiments, MoDCs were differentiated for 3 days and then 
treated with RAR siRNA or a non-specific scramble siRNA to block DC expression of
RAR. RA was added to the MoDCs 24 hours following siRNA treatment, and cells were
harvested 24 hours after RA treatment (Fig. 3E). The shorter RA exposure was used, 
because siRNA decreased DC viability. As expected, RAR siRNA reduced gene
expression of the RARA mRNA (Fig. 3F). Gene expression analysis revealed that the RA 
This article is protected by copyright. All rights reserved
i t
treatment protocol in this experiment caused a significant upregulation of ITGAE (CD103) 
gene expression in cells treated with an unspecific siRNA. Importantly, MoDCs treated 
with RAR siRNA and RA showed a significant decrease in ITGAE expression compared
with MoDCs that were treated with RA and scramble siRNA (Fig. 3G). A similar gene 
expression pattern was observed with ITGB7 (gene name of integrin 7) mRNA, although
observed changes were not significant (Fig. 3H). Interestingly, experiments with two 
different RAR inhibitors led to a paradoxical increase in CD103 expression (Suppl. Fig.
2), pointing to the involvement of additional feedback mechanisms. Overall, our data 
suggest that the RA-induced expression of CD103 in human DCs is predominantly 
mediated t RAR signaling.
TGF-1 but not RA drives phosphorylation of SMAD2/3 in human DCs.
We have previously shown that inhibition of the TGF- signaling pathway disrupts
the RA-induced increase in CD103 expression in DCs (13). However, the addition of TGF-
1 or TGF-2 during MoDC differentiation did not influence CD103 expression (13).
Similarly, concurrent exposure of MoDCs to both RA and TGF- did not lead to greater
CD103 expression than exposure to RA alone (data not shown). Several previous studies in 
other cell types indicate that there may be functional interactions between RA and TGF-
signaling pathways (13, 37). Therefore, we here evaluated whether RA could induce the 
phosphorylation of SMAD2/3 in DCs, since SMAD2/3 activation is the major signaling 
pathway downstream of the TGF- receptor. MoDCs were differentiated for 3 days in the
presence of RA, following our standard protocol. Alternatively, TGF-1, or TGF-1 and
the TGF- signaling inhibitor, SB431542, were added to cultures of differentiated MoDCs
for 30 minutes to induce SMAD2/3 activation (Fig. 4A). As expected, there was a 
significant increase in the phosphorylation of SMAD2/3 when MoDCs were cultured with 
TGF-1, which was blocked upon the addition of SB431542 (Fig. 4B, C). Interestingly, we
found no increase in the phosphorylation of SMAD2/3 when MoDCs were cultured in the 
presence of RA (Fig. 4B, C). This suggests that, consistent with our previous observations, 
This article is protected by copyright. All rights reserved
t
RA signaling in human DCs likely interacts with TGF- signaling (13) but is independent
of the SMAD pathway.
RA increases the phosphorylation of p38 MAPK in human MoDCs.
A review of the literature revealed that p38 mitogen activated protein kinase 
(MAPK) is involved in both TGF- and RA signaling pathways. Importantly, Yu et al.
demonstrated that TGF- induced p38 MAPK signaling is independent of SMAD2/3
activation (38). In addition, several recent studies have established that RA can increase the 
phosphorylation of p38 MAPK in multiple cell types (20, 21, 23, 27, 39). Based upon these 
observations and our current and previous data that demonstrated involvement of TGF-
pathways (13), but not SMAD-signaling, we investigated whether p38 MAPK plays a role 
in the RA-induced expression of CD103 on DCs. MoDCs cultured with RA for 3 days were 
harvested for Western blot analysis to investigate p38 MAPK expression and 
phosphorylation (Fig. 5A). Interestingly, we found that MoDCs cultured with RA 
demonstrated significantly increased phosphorylation of p38 MAPK compared with 
MoDCs cultured with medium alone (Fig. 5B, C). Baseline p38 MAPK expression was not 
affected by RA-treatment of the DCs (Fig. 5B). These data demonstrate that p38 MAPK is 
activated by RA signaling in human DCs.
RA-induced expression of CD103 in human MoDCs is dependent on p38 MAPK signaling.
We next determined whether p38 MAPK activation is involved in the RA-induced 
expression of CD103. To that end, MoDCs were differentiated for 3 days in the presence of 
RA and the p38 MAPK inhibitor SB202190 (Fig. 5A). When MoDCs were cultured with 
RA, we found the expected significant increase in ITGAE mRNA (Fig. 5D). However, 
MoDCs cultured with both RA and the p38 MAPK inhibitor showed ITGAE mRNA levels 
similar to baseline (Fig. 5D). In contrast, addition of the p38 inhibitor did not prevent the 
RA-induced upregulation of IL10, IL23A and IL12A and downregulation of 
This article is protected by copyright. All rights reserved
proinflammatory cytokines IL6 and TNFA (Suppl. Fig. 3), suggesting that the regulatory 
mechanisms for ITGAE gene expression are unique to this integrin. Notably, in contrast to 
monocytes treated with RA for 18 h (14), ALDH1A2 and RDH10 were not significantly 
upregulated in MoDCs differentiated in the presence of RA for 3 days (Suppl. Fig. 3), 
consistent with our previous study (13) and pointing to a transient regulation of RA 
biosynthesis pathways by RA.
We next used flow cytometry to confirm that inhibition of p38 signaling inhibits the 
RA-induced increase in CD103 protein expression. Indeed, when MoDCs were cultured 
with RA and the p38 MAPK inhibitor, the RA-induced CD103 protein expression was 
completely abrogated (Fig. 5E, F). Integrin 7 followed the same pattern of expression,
although no statistically significant difference was established (Fig. 5G). Similar trends 
were observed when geometric mean fluorescence intensity was analyzed (Suppl. Tables 1 
and 2). These data indicate that RA-induced expression of CD103 in human DCs is 
dependent on p38 MAPK signaling.
RA induces CD103 in human blood DCs through a p38 MAPK-dependent pathway
In our previous study, we were unable to demonstrate induction of CD103 expression 
in primary DCs isolated from human stomach tissue (13, 40). We hypothesized that the 
inability of gastric DCs to respond to RA was due to tissue-specific mechanisms. However, 
an alternative interpretation of these observations could be that MoDCs are not 
representative of primary human DC populations in their responsiveness to RA. To 
determine whether human DCs other than MoDCs upregulate CD103 in response to RA, 
we here used primary DCs from human blood to determine whether RA would induce 
increased CD103 expression through a p38-dependent pathway. We used magnetic beads to 
enrich CD141+ conventional DC1s (cDC1s), CD1c+ cDC2s, and CD304+ plasmacytoid 
DCs (pDCs) from peripheral human blood and treated the cells with RA in the presence or 
absence of the p38 inhibitor for 24 h (Fig. 6A, B). This exposure protocol was used, 
because blood DCs are already differentiated, but have a short life span. Importantly, both 
This article is protected by copyright. All rights reserved
cDC1s and cDC2s showed significantly increased expression of CD103 following RA 
treatment (Fig. 6C-F). This increase in CD103 expression was completely prevented by 
adding the p38 inhibitor SB202190 to the RA-treated cultures (Fig. 6C-F). A similar trend 
was observed for CD304+ pDCs; however, the percentage of CD304+ pDCs in our blood 
DCs was generally too low to provide reliable data (data not shown). These data 
demonstrate that primary human DCs respond to RA treatment by increasing CD103 
expression through a p38-dependent pathway, as also shown for MoDCs.
Inhibition of NFAT signaling prevents RA-induced upregulation of CD103 in MoDCs.  
Previous studies have described that p38 MAPK can activate the transcription factor 
nuclear factor of activated T cells (NFAT) (41, 42). Moreover, NFAT has been shown to 
interact with the enhancer region of CD103 in T cells (28). Based on these studies, we next 
asked whether NFAT activation might contribute to the induction of CD103 expression 
downstream of RARα and p38 MAPK in human DCs. MoDCs were differentiated for three
days and then were exposed to RA and various concentrations of an NFAT peptide 
inhibitor for 24 h (Fig. 7A). Importantly, as seen for the p38 inhibitor, the NFAT peptide 
inhibitor caused a dose-dependent decrease in the expression of CD103 in MoDCs cultured 
with RA that was significant at 50 µM (Fig. 7B, C and Suppl. Table 1). These data suggest 
that NFAT may be involved in the RA-induced expression of CD103 in human DCs 
downstream of RAR and p38. However, further experimental verification is needed to
link RAR p38 MAPK and NFAT signaling to CD103 expression following DC exposure
to RA.
Discussion:
Vitamin A is a key dietary micronutrient that is involved in the regulation of growth 
and development, vision and mucosal integrity and that promotes anti-inflammatory 
immune responses (1-3). In the mammalian immune system, most effects of vitamin A are 
This article is protected by copyright. All rights reserved
mediated by its major metabolite, retinoic acid (RA) (43). In a previous study that 
investigated the effect of RA on human DCs, we showed that RA drives the expression of 
CD103 (E integrin) and integrin 7 (13). However, the molecular signaling pathway that
regulates CD103 expression remains unknown. 
RA is involved in the establishment and maintenance of DC populations within the 
intestinal mucosa (4, 5, 8). Specifically, RA has an essential role in the development of pre-
mucosal DCs that give rise to intestinal DCs (5), and studies in mice showed that the 
development of the cDC2 population in the small intestinal and mesenteric lymph nodes 
(MLN) is dependent on RA signaling through RAR (4, 8). It has been well established in
human and mouse studies that RA increases RA biosynthesis in DCs through a positive 
feedback loop (12, 14). Moreover, murine CD103+ MLN DCs have higher expression of 
Aldh1a2 than CD103− DCs (2) and are potent inducers of T regulatory cells and gut homing
molecule expression (16, 44). Taken together, these data point to multiple functionally 
relevant interactions between RA production, CD103 expression, and tolerogenic DC 
function in the gut. Therefore, we here sought to define the regulatory pathways that control 
these interactions in more detail.
Our experiments revealed that RA-driven expression of CD103 is specific to DCs 
and monocytic cells. Thus, CD103 upregulation was consistently observed in MoDCs, 
regardless of differentiation stage, and also in different subsets of human blood DCs. 
Interestingly, both cDC1s and cDC2s isolated from human blood were more responsive to 
the RA than MoDCs, regardless of the RA treatment protocol used.  Monocytes cultured in 
media alone or with cytokines to differentiate into either a macrophage or DC phenotype 
showed a moderate increase in the expression of CD103 upon culture with RA. In contrast, 
differentiated monocyte-derived macrophages treated with RA showed no significant 
induction of CD103 expression. Similarly, we saw no change in CD103 expression in RA-
treated resting or proliferating CD4+ and CD8+ T cells. 
To elucidate the signaling pathway involved in RA-induced expression of CD103, 
we investigated the RA receptor (RAR) involved. RA can interact with three different 
nuclear receptors, RAR, RAR, and RAR, all of which form heterodimers with the
This article is protected by copyright. All rights reserved
retinoid X receptor (RXR) to induce gene expression upon RA stimulation (36). We here 
established that RA-induced CD103 expression is predominantly RAR-dependent.
MoDCs cultured with RA or the RAR agonist, AM80, had a 5-fold increase in CD103
expression, and to a lesser extent integrin β7. In contrast, the RAR agonist caused only a
minor increase in CD103 expression. With the RAR agonist, high dose treatment was
required to induce CD103 upregulation, which could be due to the cross-reactivity with 
RARα that the RAR agonist exhibits at higher concentrations (45). The importance of
RARα in RA-mediated CD103 expression was further confirmed through siRNA
knockdown experiments, which led to the abrogation of RA-induced CD103 expression.
RA can modify gene expression through two distinct signaling mechanisms. 
Classical genomic RA effects involve the binding of RAR to target DNA as a transcription 
factor, which can either repress or induce transcription of target genes (20-23, 46). Non-
genomic RA effects involve the activation of kinase cascades in the cytoplasm, which can 
indirectly impact gene expression through alternate pathways (20, 21, 23, 27, 39). In our 
investigations to understand the RA-induced signaling that leads to CD103 expression in 
DCs, we looked to identify non-genomic RA involvement. We previously showed that in 
the presence of RA, TGF- signaling plays a role in the expression of CD103. Inhibition of
the TGF- receptor II (TGF-RII) blocked RA-induced upregulation of CD103 in human
DCs (13). Similarly, knockout mice with a DC-specific deletion of TGF-R had fewer
CD103+ DCs and in turn a decrease in Foxp3+ T regulatory cells (47). In human 
adenocarcinoma cells, RA-induced expression of VE-cadherin was abrogated when TGF-
R was inhibite (37). Interestingly, our experiments showed that RA-induced CD103
expression was independent of SMAD2/3 activation, the major signaling pathway 
downstream of the TGF-R. P38 MAPK is a known component of the TGF- signaling,
and p APK activation by TGF- is independent of SMAD signaling (38). Therefore,
we hypothesized that p38 MAPK signaling could also be involved in RA-dependent CD103 
expression. The results from Western blot analysis of p38 MAPK support this hypothesis, 
since RA treatment lead to increased activation of p38 MAPK, as evidenced by increased 
levels of phosphorylated p38 MAPK. Notably, RA treatment of MoDCs did not alter 
This article is protected by copyright. All rights reserved
t
overall p38 MAPK expression, pointing to altered signaling pathways consistent with non-
genomic activation pathways rather than altered expression due to genomic regulation. Our 
experiments using an inhibitor of p38 MAPK strongly suggest that p38 MAPK signaling is 
involved in the induction of CD103 expression in both MoDCs and blood DC subsets, since 
p38 MAPK inhibition in RA-treated MoDCs resulted in a >2-fold decrease of CD103 
expression compared with DCs cultured with RA only. The role of p38 MAPK in RA 
signaling has previously been demonstrated in human osteoblasts and mouse embryonic 
fibroblasts ). Specifically, the additive effect of RA and TGF- on VEGF expression
in osteoblasts was highly dependent on the RA-induced p38 MAPK signaling (39). 
Additionally, in murine embryonic fibroblasts, RA-driven expression of Cyp26A1 was 
abrogated when p38 MAPK signaling was blocked (23). Thus, the results from our 
investigations with human DCs corroborate previous reports that have demonstrated the 
importance of p38 MAPK in RA signaling.
We next sought to evaluate the downstream mediators of p38 MAPK signaling. 
Multiple studies have shown that p38 MAPK can have direct interactions with the immune 
cell transcription factor NFAT (41, 42). Furthermore, investigations by Mokrani et al. 
revealed that NFAT interacts with the enhancer region of CD103 in T cells (28). Our results 
point to a role for NFAT in CD103 regulation downstream of RARα and p38 MAPK, since
increasing concentrations of NFAT inhibitor prevented RA-induced upregulation of DC 
CD103 expression in a dose-dependent manner. Future experiments will evaluate whether 
the RARα-p38-NFAT axis is the major pathway for the induction of CD103 in human
blood DCs, monocytes and MoDCs. Whether RA induction of CD103 also occurs in human 
tissues is still unclear. Thus, we were unable to show significant upregulation of CD103 in 
RA-treated human gastric DCs (13). In mice, Vitamin A deficiency leads to decreased 
numbers of CD103+ CD11b+ DCs, but not CD103+ CD11b− DCs in the intestine (8)
suggesting that additional mechanisms contribute to the regulation of CD103 downstream 
of RA.
In summary, we here have elucidated a putative signaling pathway involved in the 
RA-induced expression of CD103 in human DCs. Our data show that expression of CD103 
This article is protected by copyright. All rights reserved
in human DCs was dependent on RAR and p38 MAPK signaling. Importantly, this effect
was cell specific for MNPs, since T cells did not respond in the same way. A lack of 
sufficient nutritional RA has global health implications directly related to DC function, and 
CD103 expression has key functional correlates in mucosal immunity. Therefore, 
unraveling the DC-specific signaling pathways downstream of RA leading to CD103 
expression contributes to our overall understanding of these important molecular players. 
Acknowledgements:
Funding for this study was provided by the National Institutes of Health (NIH) 
Grants P20GM103474 (pilot funding to D.B.), R03 DK107960 and U01 EB029242 (to 
D.B.) and the USDA National Institute of Food and Agriculture Hatch Project 1015768
(D.B.); the American Association of Immunologists (AAI) Careers in Immunology 
Fellowship (M.M.R. and D.B.). We appreciate the generous support of our core facilities 
through the M.J. Murdock Charitable Trust and the NIH National Institute of General 
Medical Sciences IDeA Program P30GM110732. The authors greatly appreciate the 
support from the Montana State University Agricultural Experimental Station for the Flow 
Cytometry Core Facility at Montana State University; Drs. Mark Jutila, Jovanka Voyich, 
Edward Schmidt, and Douglas Kominsky for providing useful discussions; Andy Sebrell 
and Melissa Donovan for lab support. The authors’ sincerest thanks go to all the volunteers 
who donated blood samples for this study.
Conflict of interest:
The authors declare no conflict of interest.
Data availability:
Data availability is not applicable to this work.
This article is protected by copyright. All rights reserved
References:
1. Vermot J, Gallego Llamas J, Fraulob V, Niederreither K, Chambon P, Dolle P.
Retinoic acid controls the bilateral symmetry of somite formation in the mouse embryo. 
Science. 2005;308(5721):563-6. Epub 2005/02/26. doi: 10.1126/science.1108363. PubMed 
PMID: 15731404.
2. Huang Z, Liu Y, Qi G, Brand D, Zheng SG. Role of Vitamin A in the Immune
System. J Clin Med. 2018;7(9). doi: 10.3390/jcm7090258. PubMed PMID: 30200565; 
PMCID: PMC6162863.
3. Erkelens MN, Mebius RE. Retinoic Acid and Immune Homeostasis: A Balancing
Act. Trends Immunol. 2017;38(3):168-80. Epub 2017/01/18. doi: 10.1016/j.it.2016.12.006. 
PubMed PMID: 28094101.
4. Zeng R, Bscheider M, Lahl K, Lee M, Butcher EC. Generation and transcriptional
programming of intestinal dendritic cells: essential role of retinoic acid. Mucosal Immunol. 
2016;9(1):183-93. doi: 10.1038/mi.2015.50. PubMed PMID: 26129652; PMCID: 
PMC4698111.
5. Zeng R, Oderup C, Yuan R, Lee M, Habtezion A, Hadeiba H, Butcher EC. Retinoic
acid regulates the development of a gut-homing precursor for intestinal dendritic cells. 
Mucosal Immunol. 2013;6(4):847-56. doi: 10.1038/mi.2012.123. PubMed PMID: 
23235743; PMCID: PMC3612556.
6. Iwata M, Hirakiyama A, Eshima Y, Kagechika H, Kato C, Song SY. Retinoic acid
imprints gut-homing specificity on T cells. Immunity. 2004;21(4):527-38. Epub 
2004/10/16. doi: 10.1016/j.immuni.2004.08.011. PubMed PMID: 15485630.
7. Coombes JL, Siddiqui KR, Arancibia-Carcamo CV, Hall J, Sun CM, Belkaid Y,
Powrie F. A functionally specialized population of mucosal CD103+ DCs induces Foxp3+ 
regulatory T cells via a TGF-beta and retinoic acid-dependent mechanism. J Exp Med. 
This article is protected by copyright. All rights reserved
2007;204(8):1757-64. doi: 10.1084/jem.20070590. PubMed PMID: 17620361; PMCID: 
PMC2118683.
8. Klebanoff CA, Spencer SP, Torabi-Parizi P, Grainger JR, Roychoudhuri R, Ji Y,
Sukumar M, Muranski P, Scott CD, Hall JA, Ferreyra GA, Leonardi AJ, Borman ZA, 
Wang J, Palmer DC, Wilhelm C, Cai R, Sun J, Napoli JL, Danner RL, Gattinoni L, Belkaid 
Y, Restifo NP. Retinoic acid controls the homeostasis of pre-cDC-derived splenic and 
intestinal dendritic cells. J Exp Med. 2013;210(10):1961-76. doi: 10.1084/jem.20122508. 
PubMed PMID: 23999499; PMCID: PMC3782040.
9. Stevens GA, Bennett JE, Hennocq Q, Lu Y, De-Regil LM, Rogers L, Danaei G, Li
G, White RA, Flaxman SR, Oehrle SP, Finucane MM, Guerrero R, Bhutta ZA, Then-
Paulino A, Fawzi W, Black RE, Ezzati M. Trends and mortality effects of vitamin A 
deficiency in children in 138 low-income and middle-income countries between 1991 and 
2013: a pooled analysis of population-based surveys. Lancet Glob Health. 2015;3(9):e528-
36. doi: 10.1016/S2214-109X(15)00039-X. PubMed PMID: 26275329.
10. Vlasova AN, Chattha KS, Kandasamy S, Siegismund CS, Saif LJ. Prenatally
acquired vitamin A deficiency alters innate immune responses to human rotavirus in a 
gnotobiotic pig model. J Immunol. 2013;190(9):4742-53. doi: 10.4049/jimmunol.1203575. 
PubMed PMID: 23536630; PMCID: PMC3633669.
11. Barbalho SM, Goulart RA, Batista G. Vitamin A and inflammatory bowel diseases:
from cellular studies and animal models to human disease. Expert Rev Gastroenterol 
Hepatol. 2019;13(1):25-35. Epub 2019/02/23. doi: 10.1080/17474124.2019.1543588. 
PubMed PMID: 30791845.
12. Molenaar R, Knippenberg M, Goverse G, Olivier BJ, de Vos AF, O'Toole T,
Mebius RE. Expression of retinaldehyde dehydrogenase enzymes in mucosal dendritic cells 
and gut-draining lymph node stromal cells is controlled by dietary vitamin A. J Immunol. 
2011;186(4):1934-42. doi: 10.4049/jimmunol.1001672. PubMed PMID: 21220692.
This article is protected by copyright. All rights reserved
13. Roe MM, Swain S, Sebrell TA, Sewell MA, Collins MM, Perrino BA, Smith PD,
Smythies LE, Bimczok D. Differential regulation of CD103 (alphaE integrin) expression in 
human dendritic cells by retinoic acid and Toll-like receptor ligands. J Leukoc Biol. 
2017;101(5):1169-80. doi: 10.1189/jlb.1MA0316-131R. PubMed PMID: 28087652; 
PMCID: PMC5380378.
14. Bimczok D, Kao JY, Zhang M, Cochrun S, Mannon P, Peter S, Wilcox CM,
Monkemuller KE, Harris PR, Grams JM, Stahl RD, Smith PD, Smythies LE. Human 
gastric epithelial cells contribute to gastric immune regulation by providing retinoic acid to 
dendritic cells. Mucosal Immunol. 2015;8(3):533-44. doi: 10.1038/mi.2014.86. PubMed 
PMID: 25249167; PMCID: PMC4372513.
15. Watchmaker PB, Lahl K, Lee M, Baumjohann D, Morton J, Kim SJ, Zeng R, Dent
A, Ansel KM, Diamond B, Hadeiba H, Butcher EC. Comparative transcriptional and 
functional profiling defines conserved programs of intestinal DC differentiation in humans 
and mice. Nat Immunol. 2014;15(1):98-108. Epub 2013/12/03. doi: 10.1038/ni.2768. 
PubMed PMID: 998.
16. Jaensson E, Uronen-Hansson H, Pabst O, Eksteen B, Tian J, Coombes JL, Berg PL,
Davidsson T, Powrie F, Johansson-Lindbom B, Agace WW. Small intestinal CD103+ 
dendritic cells display unique functional properties that are conserved between mice and 
humans. J Exp Med. 2008;205(9):2139-49. doi: 10.1084/jem.20080414. PubMed PMID: 
18710932; PMCID: PMC2526207.
17. Iliev ID, Spadoni I, Mileti E, Matteoli G, Sonzogni A, Sampietro GM, Foschi D,
Caprioli F, Viale G, Rescigno M. Human intestinal epithelial cells promote the 
differentiation of tolerogenic dendritic cells. Gut. 2009;58(11):1481-9. doi: 
10.1136/gut.2008.175166. PubMed PMID: 19570762.
18. Bene K, Varga Z, Petrov VO, Boyko N, Rajnavolgyi E. Gut Microbiota Species
Can Provoke both Inflammatory and Tolerogenic Immune Responses in Human Dendritic 
Cells Mediated by Retinoic Acid Receptor Alpha Ligation. Front Immunol. 2017;8:427. 
This article is protected by copyright. All rights reserved
Epub 2017/05/02. doi: 10.3389/fimmu.2017.00427. PubMed PMID: 28458670; PMCID: 
PMC5394128.
19. Jaensson-Gyllenback E, Kotarsky K, Zapata F, Persson EK, Gundersen TE,
Blomhoff R, Agace WW. Bile retinoids imprint intestinal CD103+ dendritic cells with the 
ability to generate gut-tropic T cells. Mucosal Immunol. 2011;4(4):438-47. Epub 
2011/02/04. doi: 10.1038/mi.2010.91. PubMed PMID: 853; PMCID: 3130189.
20. Canon E, Cosgaya JM, Scsucova S, Aranda A. Rapid effects of retinoic acid on
CREB and ERK phosphorylation in neuronal cells. Mol Biol Cell. 2004;15(12):5583-92. 
Epub 2004/09/17. doi: 10.1091/mbc.e04-05-0439. PubMed PMID: 15371543; PMCID: 
PMC532036.
21. Piskunov A, Rochette-Egly C. A retinoic acid receptor RARalpha pool present in
membrane lipid rafts forms complexes with G protein alphaQ to activate p38MAPK. 
Oncogene. 2012;31(28):3333-45. doi: 10.1038/onc.2011.499. PubMed PMID: 22056876.
22. Balmer JE, Blomhoff R. Gene expression regulation by retinoic acid. J Lipid Res.
2002;43(11):1773-808. Epub 2002/10/29. doi: 10.1194/jlr.r100015-jlr200. PubMed PMID: 
12401878.
23. Bruck N, Vitoux D, Ferry C, Duong V, Bauer A, de The H, Rochette-Egly C. A
coordinated phosphorylation cascade initiated by p38MAPK/MSK1 directs RARalpha to 
target promoters. EMBO J. 2009;28(1):34-47. doi: 10.1038/emboj.2008.256. PubMed 
PMID: 19078967; PMCID: PMC2633082.
24. Chawla A, Repa JJ, Evans RM, Mangelsdorf DJ. Nuclear receptors and lipid
physiology: opening the X-files. Science. 2001;294(5548):1866-70. doi: 
10.1126/science.294.5548.1866. PubMed PMID: 11729302.
25. Dilworth FJ, Chambon P. Nuclear receptors coordinate the activities of chromatin
remodeling complexes and coactivators to facilitate initiation of transcription. Oncogene. 
This article is protected by copyright. All rights reserved
2001;20(24):3047-54. Epub 2001/06/23. doi: 10.1038/sj.onc.1204329. PubMed PMID: 
11420720.
26. Brelivet Y, Rochel N, Moras D. Structural analysis of nuclear receptors: from
isolated domains to integral proteins. Mol Cell Endocrinol. 2012;348(2):466-73. Epub 
2011/09/06. doi: 10.1016/j.mce.2011.08.015. PubMed PMID: 21888944.
27. Alsayed Y, Uddin S, Mahmud N, Lekmine F, Kalvakolanu DV, Minucci S, Bokoch
G, Platanias LC. Activation of Rac1 and the p38 mitogen-activated protein kinase pathway 
in response to all-trans-retinoic acid. J Biol Chem. 2001;276(6):4012-9. doi: 
10.1074/jbc.M007431200. PubMed PMID: 11060298.
28. Mokrani M, Klibi J, Bluteau D, Bismuth G, Mami-Chouaib F. Smad and NFAT
pathways cooperate to induce CD103 expression in human CD8 T lymphocytes. J 
Immunol. 2014;192(5):2471-9. doi: 10.4049/jimmunol.1302192. PubMed PMID: 
24477908.
29. Cepek KL, Shaw SK, Parker CM, Russell GJ, Morrow JS, Rimm DL, Brenner MB.
Adhesion between epithelial cells and T lymphocytes mediated by E-cadherin and the alpha 
E beta 7 integrin. Nature. 1994;372(6502):190-3. doi: 10.1038/372190a0. PubMed PMID: 
7969453.
30. Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image
analysis. Nat Methods. 2012;9(7):671-5. PubMed PMID: 22930834.
31. Pfaffl MW. A new mathematical model for relative quantification in real-time RT-
PCR. Nucleic Acids Res. 2001;29(9):e45. doi: 10.1093/nar/29.9.e45. PubMed PMID: 
11328886; PMCID: PMC55695.
32. Rosenzwajg M, Tailleux L, Gluckman JC. CD13/N-aminopeptidase is involved in
the development of dendritic cells and macrophages from cord blood CD34+ cells. Blood. 
2000;95(2):453-60. doi: 10.1182/blood.V95.2.453.
This article is protected by copyright. All rights reserved
33. El-Asady R, Yuan R, Liu K, Wang D, Gress RE, Lucas PJ, Drachenberg CB,
Hadley GA. TGF-{beta}-dependent CD103 expression by CD8(+) T cells promotes 
selective destruction of the host intestinal epithelium during graft-versus-host disease. J 
Exp Med. 2005;201(10):1647-57. Epub 2005/05/18. doi: 10.1084/jem.20041044. PubMed 
PMID: 15897278; PMCID: PMC2212926.
34. Liu Y, Lan Q, Lu L, Chen M, Xia Z, Ma J, Wang J, Fan H, Shen Y, Ryffel B, Brand
D, Quismorio F, Liu Z, Horwitz DA, Xu A, Zheng SG. Phenotypic and functional 
characteristic of a newly identified CD8+Foxp3−CD103+ regulatory T cells. Journal of
Molecular Cell Biology. 2013;6(1):81-92. doi: 10.1093/jmcb/mjt026.
35. Robinson PW, Green SJ, Carter C, Coadwell J, Kilshaw PJ. Studies on
transcriptional regulation of the mucosal T-cell integrin alphaEbeta7 (CD103). 
Immunology. 2001;103(2):146-54. Epub 2001/06/20. doi: 10.1046/j.1365-
2567.2001.01232.x. PubMed PMID: 11412301; PMCID: PMC1783235.
36. Bastien J, Rochette-Egly C. Nuclear retinoid receptors and the transcription of
retinoid-target genes. Gene. 2004;328:1-16. doi: 10.1016/j.gene.2003.12.005. PubMed 
PMID: 15019979.
37. Prahalad P, Dakshanamurthy S, Ressom H, Byers SW. Retinoic acid mediates
regulation of network formation by COUP-TFII and VE-cadherin expression by TGFbeta 
receptor kinase in breast cancer cells. PLoS One. 2010;5(4):e10023. Epub 2010/04/14. doi: 
10.1371/journal.pone.0010023. PubMed PMID: 20386594; PMCID: PMC2850308.
38. Yu L, Hebert MC, Zhang YE. TGF-beta receptor-activated p38 MAP kinase
mediates Smad-independent TGF-beta responses. EMBO J. 2002;21(14):3749-59. doi: 
10.1093/emboj/cdf366. PubMed PMID: 12110587; PMCID: PMC126112.
39. Tokuda H, Hatakeyama D, Akamatsu S, Tanabe K, Yoshida M, Shibata T, Kozawa
O. Involvement of MAP kinases in TGF-beta-stimulated vascular endothelial growth factor
This article is protected by copyright. All rights reserved
synthesis in osteoblasts. Arch Biochem Biophys. 2003;415(1):117-25. doi: 10.1016/s0003-
9861(03)00225-x. PubMed PMID: 12801520.
40. Roe MM, Swain S, Hashimi M, Woo KM, Bimczok D. Retinoic Acid Induces
CD103 Expression in Human Dendritic Cells through a p38 MAPK- and NFAT-dependent 
Signaling PathwayIn revision.
41. Wu CC, Hsu SC, Shih HM, Lai MZ. Nuclear factor of activated T cells c is a target
of p38 mitogen-activated protein kinase in T cells. Mol Cell Biol. 2003;23(18):6442-54. 
doi: 10.1128/mcb.23.18.6442-6454.2003. PubMed PMID: 12944472; PMCID: 
PMC193716.
42. Matsumoto M, Kogawa M, Wada S, Takayanagi H, Tsujimoto M, Katayama S,
Hisatake K, Nogi Y. Essential role of p38 mitogen-activated protein kinase in cathepsin K 
gene expression during osteoclastogenesis through association of NFATc1 and PU.1. J Biol 
Chem. 2004;279(44):45969-79. Epub 2004/08/12. doi: 10.1074/jbc.M408795200. PubMed 
PMID: 15304486.
43. Larange A, Cheroutre H. Retinoic Acid and Retinoic Acid Receptors as Pleiotropic
Modulators of the Immune System. Annu Rev Immunol. 2016;34:369-94. doi: 
10.1146/annurev-immunol-041015-055427. PubMed PMID: 27168242.
44. Guilliams M, Ginhoux F, Jakubzick C, Naik SH, Onai N, Schraml BU, Segura E,
Tussiwand R, Yona S. Dendritic cells, monocytes and macrophages: a unified 
nomenclature based on ontogeny. Nat Rev Immunol. 2014;14(8):571-8. doi: 
10.1038/nri3712. PubMed PMID: 25033907; PMCID: PMC4638219.
45. Bernard BA, Bernardon JM, Delescluse C, Martin B, Lenoir MC, Maignan J,
Charpentier B, Pilgrim WR, Reichert U, Shroot B. Identification of synthetic retinoids with 
selectivity for human nuclear retinoic acid receptor gamma. Biochem Biophys Res 
Commun. 1992;186(2):977-83. Epub 1992/07/31. doi: 10.1016/0006-291x(92)90842-9. 
PubMed PMID: 1323296.
This article is protected by copyright. All rights reserved
46. Manicassamy S, Pulendran B. Retinoic acid-dependent regulation of immune
responses by dendritic cells and macrophages. Semin Immunol. 2009;21(1):22-7. doi: 
10.1016/j.smim.2008.07.007. PubMed PMID: 18778953; PMCID: PMC2642591.
47. Bain CC, Montgomery J, Scott CL, Kel JM, Girard-Madoux MJH, Martens L,
Zangerle-Murray TFP, Ober-Blobaum J, Lindenbergh-Kortleve D, Samsom JN, Henri S, 
Lawrence T, Saeys Y, Malissen B, Dalod M, Clausen BE, Mowat AM. TGFbetaR 
signalling controls CD103(+)CD11b(+) dendritic cell development in the intestine. Nat 
Commun. 2017;8(1):620. doi: 10.1038/s41467-017-00658-6. PubMed PMID: 28931816; 
PMCID: PMC5607002.
Figure Legends:
Figure 1: RA induces increased expression of CD103 in monocytes and 
monocyte-derived DCs, but not in differentiated monocyte-derived macrophages. (A) 
Schematic of experimental design for panels B and C. Isolated CD14+ monocytes were 
cultured for 1 day with or without RA.  IL-4 and GM-CSF were added to differentiate 
monocytes into monocyte-derived DCs (MoDCs), and M-CSF was added to differentiate 
monocytes into monocyte-derived macrophages (MDMs). (B) FACS analysis of CD103 
expression on monocytes cultured under different conditions; pooled data from 3 
independent experiments; individual datapoints, mean ± SD are shown. (C) Representative 
histograms of RA-treated and untreated cells from panel B. (D) Schematic of experimental 
design for panels E and F. CD14+ monocytes were cultured for 3 days with IL-4/GM-CSF 
or with M-CSF. On day 3, RA was added to the culture for 1 day and cells were harvested 
for FACS analysis. (E) Pooled FACS data from 3-4 independent experiments showing 
CD103 expression on different cell types. (F) Representative histograms of data in panel E. 
Isotype control – grey histogram, CD103 Ab – black solid line. One-way ANOVA with 
Dunnett’s multiple comparisons test; *P < 0.05. **P < 0.01.
This article is protected by copyright. All rights reserved
Figure 2: RA does not induce significant CD103 expression in blood T cells. (A) 
Schematic of experimental design for panels B-D. CD14-depleted PBMCs were cultured 
for 1 day with or without RA, TGF-β1, or TGF-β2 and harvested for FACS analysis. (B)
Gating schematic for analysis of CD8+ and CD4+ T cells. (C-D) Pooled data from 4 
independent experiments show (C) no CD103 expression on CD4+ T cells and (D) low 
CD103 expressi  on CD8+ T cells with or without addition of RA, TGF-β1, or TGF-β2.
(E) Schematic of experimental design for panels F-I. PBMCs remaining after CD14+
monocyte depletion were cultured for 5 days with or without RA and anti-CD3/CD28 beads 
and were harvested for FACS analysis. (F) Pooled data from 3 experiments and (G) 
representative FACS histograms showing the percentage of proliferating T cells after each 
treatment as measured by CellTrace Violet for CD4+ T cells (top graphs) and CD8+ T cells 
(bottom graphs). Two-way ANOVA with Dunnet’s multiple comparisons test; *indicates 
statistical significance (P < 0.01). (H) Pooled FACS data from 3 experiments (with mean ± 
SD) and (I) representative dotplots showing CD103 expression on proliferating CD4+ and 
CD8+ T cells after 5 days in culture with or without RA treatment. *
Figure 3: RAR-α mediates RA-induced expression of CD103 on MoDCs. (A) Schematic
of experimental design for panels B-D. MoDCs were cultured with RA, AM80 (RAR-α
agonist), CD2314 (RAR-β agonist), or CD437 (RAR-γ agonist) before harvest for FACS
analysis after 3 days. RAR agonists were added at concentrations of 1 µM or 10 µM. 
Expression of (B) CD103 and (C) β7 on MoDCs was analyzed by FACS. Pooled data from
3 independent experiments; mean ± SD; one-way ANOVA with Tukey’s multiple 
comparisons test.  Symbols indicate significant difference compared to medium control; 
*P<0.05; **P<0.01; ***P<0.001. (D) Representative data showing CD103 expression on
MoDCs with RA and RAR agonists. Isotype control – grey histogram, CD103 Ab – black 
solid line. (E) Schematic of experimental design for panels F-H. MoDCs were 
differentiated for 3 days before addition of scramble siRNA or RAR-α siRNA. On day 4,
RA was added to the culture medium for 1 day before harvest and analysis by RT-qPCR. 
This article is protected by copyright. All rights reserved
i
(F) Relative gene expression of RARA, the gene encoding for RAR-α, after treatment with
RAR-α siRNA; representative experiment with n=2 technical replicates. Relative gene
expression of (G) ITGAE, gene encoding CD103, and (H) ITGB7, gene encoding β7, of
MoDCs after treatment with RA and RAR-α siRNA. RT-qPCR analyzed using the Pfaffl
method (31) and normalized to GAPDH. Pooled data from 3 independent experiments; 
mean ± SD. * reveals statistically significant difference from media control (P < 0.05). 
One-way ANOVA with Tukey’s multiple comparisons test.
Figure 4: RA does not significantly increase the phosphorylation of SMAD2/3 in 
MoDCs. (A) Schematic of experimental design. MoDCs were cultured with or without RA 
for 3 days. TGF-β1 or TGF-β1 plus SB431542, a TGF-β1 inhibitor, were added to the cell
culture, and cells were harvested for FACS analysis of SMAD2/3 phosphorylation 2 h later. 
(B) FACS analysis of SMAD2/3 phosphorylation in MoDCs, pooled data from 3
independent experiments; mean ± SD. (C) Representative FACS histograms of SMA2/3 
phosphorylation. Isotype control – grey shaded, SMAD2/3 Ab – solid black line. ** 
indicates statistical significance from the medium only control, P < 0.01. Kruskal-Wallis 
one-way ANOVA with Dunn’s multiple comparisons test.
Figure 5: RA-induced CD103 expression in MoDCs is dependent on p38 MAPK 
signaling. (A) Schematic of experimental design. MoDCs were cultured for 3 days with 
RA or SB202190, a p38 MPAK inhibitor, before harvest for Western blot, PCR, or FACS 
analysis. (B) Representative Western blot of MoDCs cultured with RA. (C) Band intensity 
of phosphorylated p38 protein normalized to GAPDH and total p38 MAPK protein. Pooled 
data from 3 independent experiments; mean ± SD. (D) Relative ITGAE gene expression of 
MoDCs cultured with RA and SB202190. Gene expression was analyzed by TaqMan qRT-
PCR, data were analyzed using the Pfaffl method. Pooled data from 3 independent 
experiments; mean ± SD.  FACS analysis of CD103 (E, F) expression and (G) β7
expression in MoDCs treated with RA and/or SB202190 (E) Pooled data (n=3) and (F) 
This article is protected by copyright. All rights reserved
t
representative data for CD103 expression. Isotype control – grey histogram, CD103 Ab – 
black solid line. (G) Pooled data for β7 expression; n=4.  *represents statistically
significant difference from medium control at P < 0.05. One-way ANOVA with Tukey’s 
multiple comparisons test.
Figure 6: RA-induced expression of CD103 on blood DC subsets is dependent on p38 
MAPK signaling. (A) Schematic of experimental design. Blood DCs were isolated for 
healthy adult donors and cultured for 1 day with the addition of RA and SB202190, a p38 
MAPK inhibitor, before harvest and FACS analysis. (B) FACS gating strategy of human 
blood DC subsets; CD141 – cDC1s, CD1c – cDC2s.  Representative FACS histograms of 
CD103 expression on (C) cDC1s and (D) cDC2s.  Isotype control – grey histogram, CD103 
antibody – black solid line. (E, F) Pooled FACS data showing CD103 expression by 
CD141+ cDC1s and CD1c+ cDC2s.  Individual data, mean ± SD, n=5. One-way ANOVA 
with Tukey’s multiple comparisons test; *P ≤ 0.05, **P ≤ 0.01, **P ≤ 0.001.
Figure 7: Inhibition of NFAT1c abrogates RA-induced expression of CD103 on 
MoDCs.  (A) Schematic of experimental design. MoDCs were differentiated for 3 days 
before the addition of RA and 10 - 50 µM NFAT inhibitor. MoDCs were harvested 1 day 
after addition of RA and NFAT inhibitor and CD103 expression was analyzed by FACS. 
(B) Expression of CD103 on MoDCs treated with RA and different concentrations of
NFAT inhibitor by FACS analysis; n=4 independent experiments; mean ± SD. (C) 
Representative FACS histograms of CD103 expression. Isotype control – grey shaded 
histogram, CD103 Ab – black solid line.  Indicates statistical significance. One-way 
ANOVA with Tukey’s multiple comparisons test, *P ≤ 0.05
Supplemental Figure 1: One-day RA-treatment of monocytes, MoDCs and MDMs 
increases expression of CD14, but does not alter HLA-DR or CD13 expression. (A) 
Schematic of experimental design for panels B, C and D. Freshly isolated CD14+ 
This article is protected by copyright. All rights reserved
monocytes or differentiated MoDCs or MDMs were cultured for 1 day with or without RA. 
Bar charts show pooled FACS analysis data (mean ± SD) from 3-4 independent 
experiments for (B) CD14, (C) CD13 and (D) HLA-DR.  Two-way ANOVA with Tukey’s 
multiple comparisons test. 
Supplemental Figure 2: Inhibitors of RARα increase the expression of CD103 in
MoDαCs. (A) Schematic of experimental design. MoDCs were cultured with RA and a 
RAR  antagonist for 3 days before harvest for FACS analysis. (B) CD103 after MoDCs
were treated with RA and different concentrations of BMS195614 (BMS), a RARα
antagonist. (C) FACS histograms of CD103 expression after treatment of MoDCs with RA 
and BMS195614. (D) CD103 expression of MoDCs treated with RA and Ro41-5253, a 
RARα antagonist. (E) FACS histograms of MoDCs treated with RA and Ro41-5253.
Isotype control – grey shaded, CD103 Ab – black solid line. Statistically significant 
difference from the medium only control: *P<0.05; **P<0.01; statistically significant 
difference from RA-treated cells: # P<0.05; One-way ANOVA.
Supplemental Figure 3: MoDC gene expression after treatment with RA and p38 
inhibitor. (A) Schematic of experimental design. MoDCs were cultured with RA and 
SB202190, a p38 MAPK inhibitor, for 3 days before harvest and gene analysis by RT-
qPCR. Relative gene expression of (B) ALDH1A2, (C) RDH10, (D) IL10, (E) IL12A, (F) 
IL6, (G) IL23A, and (H) TNFA. RT-qPCR analyzed using the Pfaffl method (31) and 
normalized to GAPDH. *indicates statistical significance at P < 0.05. ** indicates 
statistical significance at P < 0.01; one-way ANOVA. 
This article is protected by copyright. All rights reserved
imm_13246_f1.tif
This	article	is	protected	by	copyright.	All	rights	reserved
imm_13246_f2.tif
This	article	is	protected	by	copyright.	All	rights	reserved
imm_13246_f3.tif
This	article	is	protected	by	copyright.	All	rights	reserved
imm_13246_f4.tif
This	article	is	protected	by	copyright.	All	rights	reserved
imm_13246_f5.tif
This	article	is	protected	by	copyright.	All	rights	reserved
imm_13246_f6.tif
This	article	is	protected	by	copyright.	All	rights	reserved
imm_13246_f7.tif
This	article	is	protected	by	copyright.	All	rights	reserved