Vitamin D Inhibits IL-22 Production Through a Repressive Vitamin D Response Element in the il22 Promoter

Th22 cells constitute a recently described CD4+ T cell subset defined by its production of interleukin (IL)-22. The action of IL-22 is mainly restricted to epithelial cells. IL-22 enhances keratinocyte proliferation but inhibits their differentiation and maturation. Dysregulated IL-22 production has been associated to some inflammatory skin diseases such as atopic dermatitis and psoriasis. How IL-22 production is regulated in human T cells is not fully known. In the present study, we identified conditions to generate Th22 cells that do not co-produce IL-17 from naïve human CD4+ T cells. We show that in addition to the transcription factors AhR and RORγt, the active form of vitamin D3 (1,25(OH)2D3) regulates IL-22 production in these cells. By studying T cells with a mutated vitamin D receptor (VDR), we demonstrate that the 1,25(OH)2D3-induced inhibition of il22 gene transcription is dependent on the transcriptional activity of the VDR in the T cells. Finally, we identified a vitamin D response element (VDRE) in the il22 promoter and demonstrate that 1,25(OH)2D3-VDR directly inhibits IL-22 production via this repressive VDRE.


INTRODUCTION
Various subsets of effector CD4 + T helper (Th) cells, classified by the lineage-specific master transcription factors they express and the cytokines they secrete, have been described (1,2). Th22 cells constitute a recently described CD4 + T cell subset defined by their production of interleukin (IL)-22 (3,4). The biological functions of IL-22 is mainly restricted to non-hematopoietic cells such as epithelial cells located in the skin, gut, lung, liver, pancreas and kidney (5)(6)(7). In the skin and gut epithelium, IL-22 induces secretion of several anti-microbial peptides that contribute to the defence mechanisms against microorganisms (8)(9)(10). Furthermore, IL-22 enhances proliferation of keratinocytes while inhibiting their differentiation and maturation, implying an important role of IL-22 in the homeostasis of the skin (11)(12)(13). The role of IL-22 in epithelial homeostasis is further underlined by its association to inflammatory skin and gut diseases such as atopic dermatitis (14,15), psoriasis (16) and colorectal cancers (17). Th22 cells are closely related to Th17 cells, and Th17/ Th22 cells co-producing IL-17A and IL-22 have been described (10,18,19). However, distinct IL-22-producing Th22 cells that do not produce IL-17 have been isolated from both humans (3,4,20) and mice (21).
Other types of immune cells than Th22 cells, such as innate lymphoid cells (ILC), gd T cells and natural killer (NK) cells can produce IL-22 (22)(23)(24). It has been found that the transcription factor RORgt plays an important role in the regulation of IL-22 secretion in human and mouse ILC3 (25)(26)(27)(28). Furthermore, IL-21 and the aryl hydrocarbon receptor (AhR) play regulatory roles in IL-22 production in mouse CD4 + T cells (29). However, the transcription factors involved in IL-22 regulation in human Th22 cells are still not fully known. One study has found that both RORgt and AhR are important for Th22 differentiation and IL-22 production (20), whereas another study found that RORgt is undetectable in Th22 cells (4).
The aim of this study was to determine how 1,25(OH) 2 D 3 controls IL-22 production in human T cells.
First, we established the conditions for in vitro differentiation of human naïve CD4 + T cells to Th22 cells. We found that activation of naïve CD4 + T cells with allogeneic dendritic cells (DC) in the presence of TNFa, IL-6, IL-23, IL-1b, the AhR agonist FICZ and the transforming growth factor-b (TGFb) receptor type 1 inhibitor galunisertib led to optimal generation of Th22 cells. We confirmed that the transcription factors AhR and RORgt regulate IL-22 in these cells. Importantly, we found that 1,25(OH) 2 D 3 inhibits IL-22 production in human Th22 cells. We show that 1,25(OH) 2 D 3mediated inhibition of IL-22 was not due to inhibition of AhR, RORgt or STAT-3. In contrast, we identified a novel VDRE in the il22 promoter by which 1,25(OH) 2 D 3 -VDR : RXR complexes directly represses IL-22 transcription.

T Cell Isolation and Activation
All procedures involving the handling of human samples were in accordance with the principles described in the Declaration of Helsinki and the samples were collected and analysed according to ethically approval by the Regional Ethical Committee of the Capital Region of Denmark (H-16033682). Peripheral blood mononuclear cells (PBMC) were purified from healthy donor's blood by density gradient centrifugation using Lymphoprep (Axis-Shield, Oslo, Norway). Subsequently, naive CD4 + T cells were isolated by negative selection using Easysep Human Naive CD4 + T cell Enrichment Kit (19155 Stemcell Technologies) according to the manufacturer's protocol. In short, PBMC were incubated with antibodies targeting undesired cells, and subsequently magnetic particles were used to bind undesired cells. Hereafter, these cells were retained using an EasySep Magnet (18000, Stemcell Technologies). The resulting cell population consisted of >95% naïve CD4 + T cells. The obtained cells were cultured at a concentration of 1 x 10 6 cells/ ml serum-free X-VIVO 15 medium (BE02-060F, Lonza, Verviers, Belgium), and activated with allogeneic dendritic cells (DC) in a 1:10 DC:T cell ratio or activated with Dynabeads Human T-activator CD3/CD28 (111.31D, Life Technologies, Grand Island, NY) in a 2:5 bead:T cell ratio in flat-bottomed 24 well culture plates (142475, Nunc). T cells were activated for four days at 37°C, 5% CO2 under polarizing conditions for Th0 cells (un-supplemented X-VIVO 15 medium) and Th22 cells (X-VIVO 15 medium supplemented with TNF (10 ng/ml), IL-1b (10 ng/ml), IL-6 (30 ng/ml), IL-23 (20 ng/ml), FICZ (0.3 µM) and galunisertib (10 µM)). In some experiments CH-223191, SR-2211, 1,25(OH) 2 D 3 , anti-IL-21 and recombinant human IL-21 were added at the indicated concentrations to the medium during the activation period. For kinetic experiments, naïve CD4 + T cells were cultured in flat-bottomed 24-well culture plates in X-VIVO 15 medium and activated with beads or with allogeneic dendritic cells as described above in the presence of Th22 polarizing conditions for 0-144 h.

Cell Lines
The malignant T cell line MyLa 2059 was previously established from a plaque biopsy specimen of a patient with cutaneous T cell lymphoma (CTCL) (44). 5 x 10 5 cells/ml were cultured in flatbottomed, 24-well plates in RPMI-1640 supplemented with 1% Penicillin/Streptomycin, 1% L-Glutamin and 10% heatinactivated and endotoxin-free FBS in the absence or presence of Th22 polarizing conditions and the indicated concentrations of 1,25(OH) 2 D 3 for 48 h.

RT-qPCR
mRNA levels for various targets were measured by RT-qPCR. Following cell isolation, cells were lysed in TRI reagent (T9424, Sigma Aldrich) and mixed with phase separation reagent 1bromo-3-chloropropane (B9673, Sigma Aldrich). The RNA phase was isolated and mixed with isopropanol supplemented with glycogen for RNA precipitation (10814-010, Invitrogen). The RNA pellet was then washed in RNase free 75% ethanol 3 times. cDNA was synthesized from quantified RNA using High-Capacity RNA-to-cDNA ™ Kit (4387406, Applied Biosystems) according to manufacturer's instructions. For RT-qPCR, 12.5 ng of cDNA was mixed with TaqMan ® Universal Master Mix II with UNG (4440038, Applied the target primer and RNase and DNase free water for normalization. The following target primers were used: IL-22 (Hs01574154_m1), STAT-3 (Hs01047580_m1), AhR (Hs00907314_m1), RORc (Hs01076122_m1), IL-21 (Hs00222327_m1), GAPDH (Hs99999905_m1). The platebased detection instrument LightCycler ® 480 II from Roche was used for real-time PCR amplification.

Western Blotting Analysis
For Western blotting analysis, cells were lysed with lysis buffer containing 50 mM Tris-base, 150 mM NaCl, 1 mM MgCl 2 supplemented with 1% (v/v) Triton X-100, 1 x Protease/ Phosphatase Inhibitor Cocktail (5872S, Cell Signalling Technology) and 5 mM EDTA. The lysates were vortexed for 5 seconds every 5 minutes for 25 minutes at room temperature and subsequently centrifuged at 10.000 G for 10 minutes at 4°C. Loading buffer containing lithium dodecyl sulphate (LDS) (NP0007, Life Technologies) along with reducing agent (NP0009, Life Technologies) was added and the lysates separated by electrophoresis through NuPAGE ™ 10% BisTris gels (NP0302BOX or NP0301BOX, Life Technologies). The proteins were transferred to nitrocelulose membranes (LC2001, Life Technologies). The membranes were blocked in 5% skim milk dissolved in Tris-buffered saline with 0,1% tween (TBST), washed 3 times in TBST for 3 minutes and incubated overnight at 4°C with target-specific primary antibodies (anti-STAT-3 (D1B2J) and anti-phospho-STAT3 (9145) from Cell Signaling Technology, anti-VDR (D-6), anti-AhR (sc-5579), anti-RORgt (sc-293150) and anti-GAPDH (sc-365062) from Santa Cruz Biotechnology) diluted in TBST supplemented with 5% BSA. The membranes were subsequently washed and incubated with a secondary antibody (swine anti-rabbit Ig or rabbit anti-mouse Ig (P0399 and P0260 from Dako, Glostrup, Denmark S/A), conjugated with horseradish peroxidase (HRP) and diluted in 5% skim milk. Finally, membranes were washed and exposed to ECL luminescence reagent (RPN2232, Sigma Aldrich). The corresponding signals were detected using a ChemiDocTM MP Imaging System (Bio Rad) and the software ImageLab.

Plasmids
To investigate the presence of VDRE in the il22 promoter, bioinformatics analysis of the il22 gene (HGNC : HGNC:14900) was performed using JASPAR, an open-access database of transcription factor binding profiles, where several combinations of the VDR-RXR complex binding profile on several genes are described (45). The cloning of the il22 promoter into the pMCS Tluc16 Hygro Vector expressing luciferase (88255 from ThermoFisher Scientific) was performed by using the restriction enzymes KpnI and Hind III via Invitrogen GeneArt Gene Synthesis. The generated plasmid construct contained the il22 promoter, including the putative VDRE located 2159-2173 bases upstream from the start codon, driving the expression of Tluc16 luciferase gene (IL-22-Tluc VDRE-WT) and having the antibiotic resistance genes ampicillin (AmpR) and hygromycin (HygroR).

Directed Mutagenesis
The generation of the construct IL-22-Tluc with deletion of the VDRE (IL-22-Tluc VDRE-KO) was performed using the GeneArt ™ Site-Directed Mutagenesis System (A13282, ThermoFisher Scientific) according to the manufacturer's protocol for long plasmids~10 Kb. The following primers were designed and used to remove the VDRE:

Statistical Analysis
Two-tailed, paired Student's t-tests were used to compare responses in the same group of cells treated in two different ways. Significance levels are indicated as follows: * p < 0.05; ** p < 0.01; *** p < 0.005; **** p < 0.001. Data are presented as mean values with one standard error of the mean (SEM). The number of donors as well as the number of independent experiments are indicated in the figure legends.

Differentiation of Human Th22 Cells In Vitro
Presently, there is no consensus on the conditions required for in vitro generation of human Th22 cells. One study has identified IL-1b and IL-23 as the optimal cytokine cocktail to generate Th22 cells that do not produce IL-17 (20), whereas another study found that tumour necrosis factor (TNF) and IL-6 were required for optimal Th22 cell generation (4). To establish conditions for efficient differentiation of human CD4 + T cells to Th22 cells in a physiological-like setting, we activated naïve CD4 + T cells with allogeneic DC and investigated the combinatory effect of several factors believed to induce IL-22 transcription. After 96 h of culture, we measured IL-22 and IL-17 in the supernatant. In our hands, neither the IL-1b/IL-23 nor the TNF/IL-6 combination significantly increased IL-22 production compared to untreated cells ( Figure 1A). The combination of TNF, IL-1b, IL-6, and IL-23 induced both IL-22 and IL-17 secretion ( Figure 1A). The AhR agonist FICZ and the TGF-bR inhibitor galunisertib markedly increased IL-22 secretion without stimulating IL-17 secretion. Although modestly, addition of FICZ and the cytokines to galunisertib significantly increased the secretion of IL-22 without increasing the secretion of IL-17. Thus, in our hands medium supplemented with TNF, IL-1b, IL-6, IL-23, FICZ and galunisertib (from here on termed Th22 medium) resulted in optimal differentiation of human naïve CD4 + T cells towards Th22 cells that secreted high levels of IL-22 and no IL-17 ( Figure 1A). Next, we wanted to characterize differentiation of naïve CD4 + T cells in mono-cultures versus in co-cultures with allogeneic DC and furthermore to determine the time required for efficient differentiation to Th22 cells and secretion of IL-22. To do this, we measured the IL-22 concentration at day 1-6 in the supernatants of allogeneic DC-T cell co-cultures, of mono-cultures of CD4 + T cells activated with CD3/CD28 beads and of mono-cultures of activated DC all in Th22 medium. We found that the allogeneic DC-T cell co-cultures produced significantly more IL-22 than T cells activated with CD3/CD28 beads in mono-culture ( Figure 1B). Moreover, we observed that DC in mono-culture did not secrete IL-22, underlining that IL-22 is produced by T cells. Furthermore, we found that IL-22 production reached maximum and plateaued out at 96 h in the DC-T cell cocultures ( Figure 1B). Consequently, we chose DC-T cell cocultures incubated for 96 h for the following Th22 differentiation experiments.

AhR and RORgt Regulate IL-22 Production in Human Th22 Cells
Conflicting data on the role of AhR and RORgt in IL-22 production in human Th22 cells have been presented (4,20). To determine the role of these transcription factors, we stimulated naïve CD4 + T cells with allogeneic DC in Th22 medium and increasing concentrations of the AhR antagonist CH-223191 or the RORgt antagonist SR-2211. After 96 h of culture, we measured IL-22 in the supernatant. We observed that both the AhR and the RORgt antagonist down-regulated IL-22 production (Figures 2A, B). The AhR and RORgt antagonists did not affect T cell activation or viability in the concentrations used in the present study (Supplementary Figure 2). To further explore the regulatory role of AhR on IL-22 production, we activated naïve CD4 + T cells in Th22 medium in the absence or presence of the AhR agonist FICZ and the AhR antagonist CH-223191. We found that the AhR agonist up-regulated and the AhR antagonist down-regulated IL-22 mRNA and IL-22 secretion (Figures 2C, D). Taken together, these data indicate that AhR and RORgt play key roles in the regulation of IL-22 in human Th22 cells.   Figure 1A). We found that Th22 express the VDR and that 1,25(OH) 2 Figure 3). Furthermore, we found that 1,25(OH) 2 D 3 did not affect T cell activation or viability in the concentrations used in the present study (Supplementary Figure 5).

D 3 inhibits IL-22 production in Th22 mono-cultures, supporting a direct inhibitory effect of 1,25(OH) 2 D 3 on the Th22 cells (Supplementary
To establish that the inhibitory effect of 1,25(OH) 2 D 3 on IL-22 expression and production was mediated via the VDR, we determined the effect of 1,25(OH) 2 D 3 on IL-22 in parallel in CD4 + T cells from controls and from a patient with hereditary vitamin D resistant rickets. This patient has a mutation in the DNA-binding domain of the VDR that abolishes the transcriptional activity of the VDR (46). Whereas 1,25(OH) 2 D 3 clearly down-regulated IL-22 expression and production in control T cells, it had no effect on IL-22 in T cells from the patient (Figures 3E, F). Taken together, these data demonstrated that 1,25(OH) 2 D 3 inhibits IL-22 expression and secretion in human Th22 cells and that the 1,25(OH) 2 D 3 -induced inhibition of IL-22 is dependent on the transcriptional activity of the VDR.

1,25(OH) 2 D 3 Does Not Inhibit IL-22 by Affecting AhR, RORgt or STAT-3 Expression
Recently, it has been reported that the transcription factors AhR, RORgt and STAT-3 play critical roles in IL-21-mediated induction of IL-22 in mouse T cells (29). In the present study, we found that the transcription factors AhR and RORgt regulate IL activated naïve CD4 + T cells with allogeneic DC in Th22 medium in the absence or presence of 1,25(OH) 2 D 3 . After 96 h of culture, we determined the mRNA and protein expression levels of AhR, RORgt and STAT-3. We found that the mRNA and protein expression levels of AhR ( Figures 4A, B), RORgt (Figures 4C,  D) and STAT-3 ( Figures 4E, F) were not significantly affected by 1,25(OH) 2 D 3 . Furthermore, 1,25(OH) 2 D 3 did not significantly affect the phosphorylation of STAT-3 ( Figure 4F). Thus, 1,25 (OH) 2 D 3 did not inhibit IL-22 expression and production by inhibition of AhR, RORgt or STAT-3 expression or STAT-3 phosphorylation in human Th22 cells.

IL-21 Does Not Rescue IL-22 Production in 1,25(OH) 2 D 3 -Treated Th22 Cells
In mice, IL-21 promotes IL-22 production in CD4 + T cells (29). To investigated whether 1,25(OH) 2 D 3 regulate IL-21 in human Th22 cells, we activated naïve CD4 + T cells in Th22 medium in the absence or presence of 1,25(OH) 2 D 3 and measured IL-21 concentration in the supernatant at day 4. We found that 1,25 (OH) 2 D 3 down-regulated IL-21 mRNA and protein expression in human Th22 cells ( Figure 5A and Supplementary Figure 4A). Even though we did not find IL-21 to up-regulate IL-22 in the absence of 1,25(OH)2D3 ( Figure 5B), the possibility existed that  Figures 5C, D). Likewise, we found that anti-IL-21 antibodies did not inhibit IL-22 production although it strongly neutralised IL-21 in the culture supernatants (Supplementary Figures 4A, B). We found a potential VDRE sequence in the il22 promoter located 2159-2173 base pairs upstream from the start codon of the il22 gene ( Figure 7A). To determine whether this sequence actually represented a repressive VDRE, we constructed two reporter vectors where luciferase expression was dependent on the il22 promoter. One of the vectors contained the wild-type il22 promoter sequence including the putative VDRE (IL22-TLuc VDRE-WT) and the other vector contained the il22 promoter sequence where the putative VDRE was deleted (IL22-TLuc VDRE-KO) ( Figure 7B). We transfected Myla 2059 cells with  Figure 7B). These data indicated that the il22 promoter contains a repressive VDRE located 2159-2173 base pairs upstream from the start codon of the il22 gene.

DISCUSSION
In this study, we show that 1,25(OH) 2 D 3 inhibits IL-22 expression and production in human Th22 cells through a repressive VDRE in the il22 promoter. 1,25(OH) 2 D 3 is well-known by its immunomodulatory properties and it can influence the differentiation of T helper cells by regulating the production of their signature cytokine (35)(36)(37)(38)(39)(40)(41). Some studies have investigated the effect of 1,25(OH) 2 D 3 on IL-22 in human and mice CD4 + T cells (4,42,43). However, conflicting results were obtained. One study found that 1,25(OH) 2 D 3 inhibited IL-22 production in human Th17 cells (43) In the present study, we describe a novel way to differentiate human Th22 cells in vitro. In our system, where we activate naïve CD4 + T cells with allogeneic dendritic cells, the sole presence of IL-6 and TNFa did not increase IL-22 production. However, we found that IL-6, TNFa, IL-1b and IL-23 lead to an increase in both IL-22 and IL-17 production. This is in accordance to previous studies that found that these factors increase IL-17 production in CD4 + T cells (52)(53)(54). As the AhR agonist FICZ has been found to induce IL-22 and inhibit IL-17, we included FICZ in the Th22 panel. We observed that FICZ lead to increased IL-22 production while down-regulating IL-17. However, IL-17 was still produced to some extent in the presence of FICZ. A key cytokine in the differentiation of Th17 cells is TGFb (55)(56)(57). Thus, we included an inhibitor of TGFbR signalling (galunisertib) in an attempt to repress the generation of IL-17producing CD4 + T cells in the presence of factors that induce IL-22 production. Interestingly, we found that galunisertib augmented the production of IL-22 while inhibiting IL-17 production. Taken together, we found that the combination of IL-6, TNFa, IL-1b, IL-23, FICZ and galunisertib constituted optimal conditions for in vitro generation of human Th22 cells.
We found that AhR and RORgt are important transcription factors that regulate IL-22 in human Th22 cells. In accordance, AhR and RORgt regulate IL-22 expression and production in ILC3 that represent a major IL-22 source in the gut (58,59). In contrast to a recent study that identified IL-21 as an inducer of IL-22 production in mouse CD4 + T cells (29), we found that IL-21 do not affect IL-22 production in human Th22 cells. Thus, our data suggest that regulation of IL-22 may differ between mice and human CD4 + T cells.
Interestingly, ectopic IL-22 expression is a characteristic feature of lesional skin in CTCL, and IL-22 is believed to play a role in the establishment of the pro-tumorigenic microenvironment and the deficient antimicrobial defence in these patients (47,60). Of notice, CTCL lesions are often localized to body areas, which are not exposed to sunlight i.e. the "bathing suit area" and in general, CTCL patients display deficient vitamin D serum levels (49). Given the present findings that vitamin D inhibit IL-22 expression in malignant T cells, we hypothesize that vitamin D supplementation could have a beneficial effect as adjuvant therapy inhibiting ectopic IL-22 expression and skin inflammation in CTCL.
In conclusion, we have identified a novel way of differentiating naïve CD4 + T cells towards the Th22 lineage and demonstrated that 1,25(OH) 2 D 3 directly inhibits IL-22 through VDR targeting a repressive VDRE located in the il22 gene. We showed that AhR and RORgt regulate IL-22 in human Th22 cells, whereas IL-21 does not affect IL-22 production in human Th22 cells. This study add to the understanding on IL-22 regulation in human Th22 cells and suggests that vitamin D may be considered a potential therapeutics to regulate IL-22mediated diseases.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by Regional Ethical Committee of the Capital Region of Denmark. The patients/participants provided their written informed consent to participate in this study.

AUTHOR CONTRIBUTIONS
CG, MK-W and DL conceived the study and designed the experiments. DL, FA-J, ND, UP and ST performed the laboratory experiments. CB, BW, AW and NØ assisted with the experimental design and data interpretation. CG, MK-W and DL analysed the data and wrote the manuscript with input from all authors. All authors contributed to the article and approved the submitted version.