Early Stress-Response Gene REDD1 Controls Oxazolone-Induced Allergic Contact Dermatitis

Abstract

REDD1 is an energy sensor and stress-induced mTOR inhibitor. Recently, its novel role in linking metabolism and inflammation/immune responses has emerged. Here we assessed the role of REDD1 in murine oxazolone (Oxa)-induced allergic contact dermatitis (ACD), a T cell-dependent model with features of human ACD. A variety of immune indices, including edema, cellular infiltration, inflammatory gene expression and glucocorticoid response, were compared in Redd1 knockout (KO) and isogenic (C57BL/6×129)F1 wild type mice after sensitization and subsequent ear challenge with Oxa. Despite relatively normal thymic profiles and similar T cell populations in the lymph nodes of naïve Redd1 KO mice, early T cell expansion and cytokine production was profoundly impaired after sensitization. Surprisingly, higher steady state populations of CD4⁺ and CD8⁺ T cells, as well as macrophages (CD45⁺/Ly6G⁻/CD11b⁺), dendritic cells (CD45+/Ly6G-/CD11c+), neutrophils (CD45+/Ly6G+/CD11b+) and innate lymphoid cells (ILC2s:CD45⁺/Lineage⁻/IL-7Ra⁺/ST2⁺/c-kit⁺) were observed in the ears of naïve Redd1 KO mice. Upon challenge, ear edema, T cell, macrophage, neutrophil and dendritic cell infiltration into the ear were significantly reduced in Redd1 KO animals. Accordingly, we observed significantly lower induction of IFNγ, IL-4 and other cytokines as well as pro-inflammatory factors including TSLP, IL-33, IL-1β, IL-6 and TNFα, in challenged ears of Redd1 KO mice. The response to glucocorticoid treatment, was also diminished. Taken together, these data establish REDD1 as an essential immune modulator that influences both the initiation of ACD disease, by driving naive T cell activation, and the effector phase, by promoting immune cell trafficking in T cell-mediated skin inflammation.

Introduction

REDD1 (regulated in development and DNA damage response 1), also known as DDIT4 (DNA-damage-inducible transcript 4), is an early response gene that regulates the response to a variety of cellular stresses, including deficits of cellular energy/nutrients, hypoxia, DNA damage, endoplasmic reticulum stress and viral infection (1). The best-known REDD1 function is its ability to negatively regulate mammalian target of rapamycin 1 (mTORC1), a signaling pathway that integrates cell growth and survival. It acts by stabilizing tuberous sclerosis complex 1–2 (TSC1)-TSC2 inhibitory complex (2, 3) or as shown more recently, via control of Akt^Thr308 dephosphorylation (4). Yet, some REDD1 effects, for example, modulation of nuclear factor κB (NF-κB) and hypoxia-inducible factor-1α (HIF-1α), appear to be mTOR-independent (5).

Most early studies focused on the role of REDD1 in the control of cell growth, apoptosis, and autophagy, as well as on cancer development (1, 6). However, more recent data suggests that REDD1 is also implicated in the regulation of immune responses, suggesting a contribution to the development of inflammatory diseases. Indeed, REDD1 is overexpressed in immune cells of patients with ulcerative colitis, systemic lupus erythematosus, and in the lungs of patients with emphysema (7–10). In animal studies REDD1 was shown to be induced in immune and endothelial cells after exposure to inflammatory stimuli, such as lipopolysaccharide (LPS) or cigarette smoke condensate (11–13). Moreover, Redd1 knockout animals appear to be more resistant to excessive and damaging inflammatory responses in these settings. Skin inflammatory diseases are among the most common diseases in the world. Yet, there have been no systematic analyses of the effect of REDD1 on specific immune cell responses in a skin disease model.

Allergic contact dermatitis (ACD) is a common dermatological condition that affects 15–20% of the world’s population and is characterized by local edema, rash, and itch (14). Two discrete phases are required for disease manifestations: sensitization and elicitation. Sensitization occurs upon initial skin exposure to small chemical compounds called haptens. During this phase, antigen presenting cells including resident dermal dendritic cells as well as epidermal Langerhans cells take up hapten-modified proteins, and migrate to the draining lymph nodes where they present these complexes and activate hapten-peptide specific naïve T cells to proliferate and undergo differentiation to effector T cells. Upon challenge, T cells migrate to the affected tissues and are reactivated by local exposure to hapten complexes resulting in the generation of memory T cells, cytokine and chemokine production and recruitment of additional inflammatory cells to the sites of allergen contact (14, 15). Although ACD is considered a T cell-dependent disease, keratinocytes, dendritic cells, macrophages, mast cells and neutrophils also play essential modifying roles in both the sensitization and elicitation phases (14).

Using Redd1-deficient (Redd1 KO) mice and their isogenic wild type (WT) counterparts, we assessed the role of REDD1 in murine contact dermatitis, a T cell-mediated disease model with features of allergic contact dermatitis (ACD) in human skin, induced by the topical application of the hapten, oxalozone (Oxa) (14). Despite relatively normal indices of thymic T cell development and naïve T cell residence in the draining lymph nodes of Redd1 KO mice, these mice have profound deficits in early T cell activation after sensitization. REDD1 also selectively limits steady state populations of many resident immune cell populations in the ear skin. We also show that Redd1 mRNA is induced in WT ear skin within 2 hours of challenge and is required for normal inflammatory cell infiltration and the expression of numerous inflammatory genes upon Oxa challenge. These data demonstrate the widespread influence of REDD1 in controlling T cell and innate immune cell function in ACD.

Material and methods

Reagents

Oxazolone (Oxa, 4-ethoxymethylene-2-phenyl-2-oxazolin-5-one), fluocinolone acetonine (FA), phorbol-12-myristate 13-acetate (PMA) and ionomycin were purchased from Sigma Aldrich (St. Louis, MO), Brefeldin A was purchased from Thermo Fisher Scientific (Waltham, MA). Oxa was prepared as 1–2% solution in 100% ethanol immediately prior to use. FA was dissolved in acetone and kept at −20°C until use.

Ethics Statement

All animal housing and treatments were performed in accordance to the animal protocols approved by the Northwestern University Animal Care and Use Committee, in a strict adherence to the National Institutes of Health guidelines.

Animals

All experiments were performed on adult (7–8 weeks old) female mice. Redd1 KO mice were raised in house (16). Isogenic F1 C57BL/6×129 mice were purchased from Taconic Farms (Germantown, NY) and acclimated in the same housing facility for at least 4 days before experiments.

Oxazolone sensitization and challenge

Animals were sensitized by topical application of 100 µL of vehicle (ethanol) or 2% Oxa in ethanol to the shaved abdomen. Six days later, 10 µL of 1% Oxa was applied to both sides of right ear. Left ears were treated similarly with vehicle (see Figure 1). For time course experiments, mice were challenged and sacrificed 0.5–24 hours later.

Figure 1: Redd1 is induced in contact hypersensitivity responses and promotes edema.

(A) Eight week old WT and Redd1 KO female mice were sensitized by topical application of Oxalozone (100 μL of 2% Oxa in ethanol) to shaved abdominal skin. Six days later mice were challenged by Oxa application (10 μL of 1% Oxa in ethanol) to both sides of the right ear. Left ears received ethanol vehicle alone. Immune indices in the draining lymph nodes were assessed at 3 days post sensitization and in the right and left ears at 48 hours post challenge. (B) qPCR analysis of Redd1 expression in the ears of WT mice 2 hours post challenge. Data are expressed as percent of vehicle control and presented as Mean ± SD, n=3 mice/group, **P<0.01 by two-tailed unpaired t-test, 0 vs 2 hours post-Oxa challenge. (C) Representative images of H&E staining of vehicle (control) and Oxa-challenged (Oxa) ear sections at 48 h after challenge. (D) Morphometric analysis of ear thickness, a measure of inflammation, was recorded for each section (] - indicates the measured part of the ear). Data are represented as mean ± SD of at least 10 measurements/ear, n=4 individual samples/group. (E) Inflammation measured by increase in ear weight at 24, 48 and 72 hours post challenge relative to ears of “sensitized only” (0 hr. time point mice). Data represent individual ear measurements of four animals per group. *P< 0.05 by two-tailed unpaired t-test, Redd1 KO (red) vs WT (blue).

Glucocorticoid FA effect on oxazolone challenge

Wild-type and Redd1 KO mice were sensitized with Oxa as described above. Six days later, 10 μL of fluocinolone acetonine (FA) solution in acetone or solvent control was applied to both sides of right ear one hour prior to Oxa challenge treatment.

Morphometric analysis

Mice were sacrificed 48 hours after Oxa challenge, ears were removed and fixed in 10% neutral-buffered formalin. Hematoxylin and eosin (H&E) staining was done in formalin-fixed, paraffin-embedded samples. Ear thickness was determined from the top of epidermal layer to cartilage. At least ten individual fields of view/slide in four individual ear samples (40 images/treatment group) were analyzed. H&E staining was performed at Northwestern Mouse Histology and Phenotyping Core.

Analysis of ear swelling responses by measurement of ear weights

Mouse ears were challenged with 1% Oxa as described above. Animals were sacrificed 24, 48 and 72 hrs later and ears were harvested and the weight of 5 mm ear punch, a readout of inflammation, was measured. Ears from mice that were sensitized but not challenged were used as controls.

Cell analysis by flow cytometry

Ears were harvested for flow cytometry 48 hours post Oxa challenge. Axillary, and inguinal lymph nodes were harvested 72 hours post sensitization. Thymuses were harvested from naïve mice. Tissue was homogenized and treated with collagenase IV and DNase to prepare single cell suspensions. Before staining with indicated antibodies (listed in Supplementary Table I), samples were treated with Fc block (Biolegend, San Diego, CA) in 100 µl FACS buffer for 30 minutes in the dark at 4°C. For intracellular staining, cells were stimulated with plate-bound CD3 antibodies for 5 hours and 3 μg/ml Brefeldin A was added for the final 4 hours. Cells were assayed for cytokine production using the Fixation & Permeabilization Kit (Thermo Fisher Scientific). Samples were processed on a BD FACS CantoTM II (BD Biosciences). Gating strategy is included in relevant figures as well as in Supplemental Figure I. Gates were determined based on fluorescence minus one (FMO) control staining samples. Antibodies used are shown in Supplemental Table I.

Real-time polymerase chain reaction (RT-PCR), quantitative RT-PCR (qPCR)

Mouse ears were snap frozen in liquid N₂ and kept at −80°C. Frozen ears were ground in liquid N₂ using a mortar and pestle. RNA was extracted from ear tissues using EZNA total RNA kit 1 (Omega BioTrek, Norcross, GA). Random hexamers, M-MLV reverse transcriptase (Thermo Fisher Scientific) and 1 μg of total RNA were used for reverse transcription. Gene-specific primers were designed using the NCBI Primer-BLAST tool (http://www.ncbi.nlm.nih.gov/tools/primer-blast/) and their sequences are presented in Supplementary Table II. qPCR was performed with SsoAdvanced™ Universal SYBR Green Supermix (Bio-Rad Laboratories, Hercules, CA). Fold change was calculated as 2^∆∆Ct and the values were normalized to the housekeeping Rpl27 transcript. All samples were tested in triplicate in 2–3 independent experiments.

Statistical analyses

Mean and standard deviation were calculated using Microsoft Excel software. Statistical analyses of the data were performed by non-parametric unpaired two-tailed Student’s t-test using GraphPad Prism software (version 7.03).

Results

Edema development in response to Oxa challenge is reduced in Redd1 KO mice.

To assess the effects of Redd1 on hypersensitive responses in the ear, Redd1 KO and isogenic WT mice were sensitized with Oxa by topical application of 100 µL of 2% Oxa to shaved abdominal skin. Six days later, sensitized mice were treated with vehicle alone on the left ear or challenged with Oxa on the right ear (the experimental design is presented in Figure 1A). We first evaluated expression of Redd1, an immediate/early response gene (1), in mouse ears. As shown in Figure 1B, Oxa led to significant increases in Redd1 expression 2 hours post challenge. The influence of REDD1 in ear inflammation induced by Oxa was assessed using morphometric analysis of ear thickness as a readout of edema/inflammation 48 hours after Oxa challenge. Ear-swelling responses were observed in both WT and Redd1 KO mice. However, the Oxa-induced increase in ear thickness was moderately, albeit significantly, reduced in Redd1 KO mice (Figures 1C and D).

T cell development is not impaired in Redd1 KO mice.

As discussed, the development of allergic contact dermatitis relies on efficient hapten sensitization of T cells (14). Although REDD1 is expressed in activated T cells (17), little is known regarding its role in T cell development and function. Thus, we first assessed the consequences of REDD1 deficiencies on early T cell differentiation by analyzing T cell populations in the thymus. As shown in Figures 2A and B, proportions of CD4 and CD8 single positive (SP) cells as well as CD4 CD8 double positive (DP) and double negative (DN) cells in the thymus are similar in both groups. The DN population, defined in part by distinct patterns of CD25 and CD44 expression, comprises the earliest T cell precursors that ultimately give rise to mature CD4+ and CD8+ T cells. DN1 cells are considered pluripotent and, in addition to T cells, they can generate B cells, natural killer (NK) cells, NKT cells, dendritic cells (DCs), and even myeloid cells. DN2 cell intrathymic differentiation potential is limited to T cells, ILC2s and DCs, while DN3 and DN4 cells are fully committed T cell precursors (18–20). In Redd1 KO mice there is a significant reduction in thymic DN1 cells and a significant, albeit slight, increase in DN2 cells (Figures 2C and D). However, DN3 and DN4 populations show no differences between groups, suggesting there are no major defects in thymic T cell development.

Figure 2: T cell development in the thymus is not impaired in Redd1 KO mice.

Single-cell suspensions were prepared from the thymuses of naïve WT and Redd1 KO mice. Cell populations were identified by flow cytometry after staining with indicated antibodies. (A) Representative flow cytometry plots of thymic CD45+ CD4+ CD8+ cell populations. (B) Proportions of thymic CD4 and CD8 single positive (SP) cells as well as CD4 CD8 double positive (DP) and CD4 CD8 double negative (DN) cells. (C) Representative flow cytometry plots of thymic CD45+ CD4- CD8- (DN) subset populations. (D) Proportions of thymic CD4 and CD8 double positive (DN) cells from both groups of mice. Data shown are representative of one of two independent experiments with 4 mice/experimental group.

REDD1 is required for optimal T cell activation and cytokine production in the lymph nodes of sensitized mice.

Analyses of the lymph nodes of naïve wild type and Redd1 KO mice revealed that CD4+ and CD8+ T cells are present in similar numbers (Figure 3A and B), supporting the idea that REDD1 deficiencies do not affect T cell development. However, sensitization, assessed at three days after Oxa treatment, results in a significant expansion of both CD4+ and CD8+ T cells in the draining lymph nodes of wild type mice. Yet, there was an actual reduction in total T cell numbers in KO mice (Figure 3B), consistent with a previous in vitro study showing that REDD1 is required for optimal T cell proliferation and survival in response PHA or αCD3/CD28 antibody activation (17). Not surprisingly, T cell cytokine production also appears to be compromised in these mice. Not only are T cell numbers significantly lower in the lymph nodes of sensitized Redd1 KO mice, a smaller proportion of CD4+, but not CD8+, T cells from Redd1 KO lymph nodes expressed IFNγ, IL-17 and IL-5 after in vitro reactivation with anti-CD3 antibodies (Figure 4). The mean fluorescent intensity of both CD4+ and CD8+ T cell cytokine responses was reduced as well. These results illustrate that REDD1 is essential for the early antigen-driven proliferation/expansion and cytokine responses of T cells in the lymph nodes post-sensitization.

Figure 3: CD4+ and CD8+ T cell priming is impaired in lymph nodes of Redd1 KO mice post-sensitization.

Single-cell suspensions were prepared from the ears of naïve and WT and Redd1 KO mice 48 h post sensitization. Cytokine expression was analyzed by flow cytometry after a 4 hour in vitro re-stimulation with anti-CD3 antibodies. (A) Flow cytometry gating strategy showing fluorescence minus one (FMO) controls (B) A comparison of cell numbers in the lymph nodes of naive and sensitized mice. Data shown are representative of one of three experiments.

Figure 4: Reduced T cell cytokine production in the draining lymph nodes of Redd1 KO mice post-sensitization.

Cells were prepared as described in Figure 3 and restimulated for 5 hours with anti-CD3 antibodies prior to staining with cytokine-specific antibodies. (A) Flow cytometry gating strategy showing FMO control samples. (B) Relative cytokine production depicted as % expressing cells (top row) and Mean Fluorescence Intensity (MFI) (bottom row). Data is representative of 2 independent experiments. *P<0.05, **P<0.01, by two-tailed unpaired t-test.

Redd1 KO mice display an increased number of CD45⁺ cells at steady-state in ear skin but a reduced response to Oxazolone challenge.

To assess the effects of REDD1 on local T cell infiltration in response to Oxa challenge, we compared populations of total CD45+ cells, T cells and type 2 innate lymphoid cells (ILC2s: Lineage-, IL-7Rα+, ST2+) in the ears of naïve and sensitized/challenged mice using flow cytometry analysis (Gating strategy for ILC2s is shown in Supplementary Figure I). Surprisingly, naïve Redd1 KO animals (non-sensitized, non-challenged) displayed increased proportions of CD45⁺ cells at steady state (Figure 5A). CD45⁺ populations in WT ears comprised ~1.78% of the total cells, while the mean CD45⁺ population in Redd1 KO ears was 8.45%, ~4 times higher (P<0.0001). This increased cellularity was reflected in significantly increased basal populations of CD4+ T cells, CD8+ T cells and ILC2s in Redd1 KO ears compared to wild type ears.

Figure 5: Redd1 KO mice display an increased number of T cells at steady-state in ear skin but a reduced response to Oxalozone challenge.

Single-cell suspensions were prepared from the ears of naïve, vehicle-treated (sensitized only) - and Oxa-challenged WT and Redd1 KO mice 48 h post challenge. Cell lineages were identified by flow cytometry. (A) Proportions of CD45+, CD4+ and CD8+ T (gated on CD45+/CD3+ cells) cells and ILC2s (CD45+, Lineage-, IL7Ra+, ST2+) in the ears of untreated mice. (B) Innate immune cell increases in response to Oxa challenge. Data are presented as fold increase relative to ears of naïve mice of corresponding genotype. Data are compiled from 2–3 independent experiments. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 by two-tailed unpaired t-test.

We next compared these same populations in vehicle-treated left ears (sensitized only) and Oxa-treated right ears. To account for differences in baseline immune cell populations, data are expressed as fold increase relative to naïve mice from the same group. As shown in Figure 5B, while both WT and KO mice exhibit significant CD45+ cell infiltration to the challenged ear relative to vehicle only ear, the magnitude of infiltration is higher in WT mice (average 16.1-fold increase in WT ears versus 8.3 in Redd1 KO ears, p<0.0001). The difference in the infiltration of T cells in wild type and Redd1 KO mice was most striking. An average 12-fold increase in CD4+ T cells and an average 25-fold increase in CD8+ T cells was observed in the Oxa-challenged WT ears, whereas both the CD4+ and CD8+ T cell response was negligible in challenged Redd1 KO animals. (p<0.01). Thus, REDD1 plays a seminal role in facilitating T cell migration to the affected ear upon Oxa challenge.

Group 2 innate lymphoid cells (ILC2s, nuocytes) are Lineage negative, ST2+, c-kit (CD117)+, IL-7Rα+ (CD127) immune cells that have been implicated in a TNCB-induced model of contact hypersensitivity (21) and Oxa-induced colitis (22). Skin-derived ILC2s express ST2, the IL-33 receptor, and are enriched in skin lesions of atopic dermatitis patients (23). Although we observed ILC2s in the ears of naïve animals of both genotypes, Oxa challenge did not significantly increase ILC2s in either WT or Redd1 KO mice (Figures 5A and B).

These patterns are also evident in non-lymphoid cells. Macrophages (CD11b+, Ly6G-, CD11c), dendritic cells (CD11c+, Ly6G-) and neutrophils (CD11b+, Ly6G+) exhibit significant increases in steady state ear populations in Redd1 KO mice relative to wild type animals. Only mast cells (FcεR1+, c-kit+) show no differences between groups while basophils are decreased in KO animals (Figure 6A).

Figure 6: Redd1 KO mice display an increased number of innate immune cells in ear skin at steady-state but a reduced response to Oxalozone challenge.

Single-cell suspensions were prepared from the ears of naïve, vehicle-treated (sensitized only) - and Oxa-challenged WT and Redd1 KO mice 48 h post challenge. Cell lineages were identified by flow cytometry. (A) Proportions of innate immune cells in the ears of untreated mice. Macrophages: Ly6G⁻/CD11b+, Mast cells: c-kit+/FcεR1+, DCs: Ly6G-/ CD11c+/CD11b+/−, Neutrophils: Ly6G+/CD11b+, Basophils: c-kit-/FcεR1+, ILC2s: Lineage/IL7Rα⁺/ST2⁺. Data represents the compilation of 2–3 independent experiments. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 by two-tailed unpaired t-test. (B) Cell increases in response to Oxa challenge. Data are presented as fold increase relative to ears of naïve mice of corresponding genotype. Data are compiled from 2–3 independent experiments. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 by two-tailed unpaired t-test.

In WT mice, these same innate immune cell populations show apparent increases in the challenged ear in response to Oxa treatment, but only macrophages, neutrophils, basophils and mast cells show significant fold increases over vehicle-treated ears (Figure 6B). There is also a significant Oxa-induced infiltration of macrophages, neutrophils and mast cells in Redd1 KO mice. However, with the exception of mast cells, the fold increase in KO mice is diminished in all cell types compared to WT. For example, Oxa challenge resulted in an average 4.2-fold increase in ear macrophages in WT animals compared to only a 2.4-fold increase in Redd1 KO mice (p<0.01). Neutrophils are increased by an average of 12-fold in WT ears but only ~4.8 -fold in Redd1 KO ears (p<0.0001).

Cytokine and chemokine expression are reduced in Redd1 KO mice after Oxa challenge.

We next analyzed the expression of a variety of inflammatory genes in Oxa-challenged ears at 24 hours post challenge (Figure 7). The induction of several cytokine genes including Il1b, Il6, Tnf, Ifng Il33 and Il4 are reduced in Redd1 KO mice relative to WT mice, with Il6, Tnf, Ifng and Il33 exhibiting the most dramatic differences. The chemokine receptors, Cxcr3 and Ccr4, both of which regulate T cell migration, as well as Cxcl10, a chemoattractant for macrophages and dendritic cells, also show impaired expression relative to WT mice.

Figure 7: Reduced inflammatory gene expression in Oxa-challenged Redd1 KO mice.

Total RNA was isolated from mouse ears 24 hours post Oxa challenge and gene expression was assessed by qPCR analyses. Results were normalized to the expression of housekeeping gene Rpl27. Hprt1 was also used to normalize a subset of data with similar results. Data (n=3–6 mice/group) are presented as Mean ± SD. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 by two-tailed unpaired t-test. Data shown are representative of two independent experiments with 3–6 mice/experimental group.

In line with reduced Il1b expression, the inflammasome genes Nlrp3 and Casp1 are significantly reduced in expression as is Nox1, which regulates membrane associated ROS production. The expression of two genes associated with ILC2 effector function, Tslp (thymic stromal lymphopoietin) and Areg (amphiregulin) is lower in Redd1 KO ears as well. Il2 and Il9 show a reduction in Redd1 KO mice relative to WT values, but these differences do not reach statistical significance.

Glucocorticoids exhibit reduced anti-ACD activity in Redd1 KO mice

Topical glucocorticoids are widely used for the treatment of inflammatory skin diseases including ACD [20, 26]. Thus, we asked whether Redd1 status alters the anti-inflammatory effects of the glucocorticoid, fluocinolone acetonide (FA; class III potency glucocorticoid widely used in dermatology), in the Oxa-ACD model. FA (0.025 and 0.25 µg/animal) was topically applied to the ear 1 hour before Oxa challenge. Cellular infiltration was analyzed 48 hours post-challenge and cytokine gene expression at 24 hours. In WT mice, FA treatment reduced ear infiltration of total CD45⁺ cells by two-fold (Figure 8A), and drastically decreased induction of Ifng, Tnf, IL1b, Il6, Il4 and Il33 (Figure 8B). The anti-inflammatory effects of FA, especially at the higher dose, were significantly reduced in Redd1 KO mice.

Figure 8: Redd1 KO mice show reduced sensitivity to the anti-inflammatory effects of glucocorticoid fluocinolone acetonide (FA).

Eight-week old female WT and Redd1 KO mice were sensitized by topical application of Oxa (100 μL of 2% Oxa in ethanol) as depicted in Figure 1. Six days later, right ears were pre-treated with solvent control (acetone) or either 25 ng or 250 ng FA and one hour later, challenged by application of 1% Oxa (10 μL). (A) Single-cell suspensions were prepared from Oxa-challenged right ears (pre-treated with solvent or 250 ng FA) of WT and Redd1 KO mice 48 h after the challenge. Cells were analyzed by flow cytometry using dye-conjugated CD45-specific antibodies, as described in Materials and Methods. Data is representative of two experiments with 3–4 mice/group. (B) Total RNA was isolated from right ears 24 hours after Oxa challenge. Q-PCR analyses were performed using mouse-specific primers. Data is presented as percent of solvent control-challenged ears (no FA) of parent control group, which was set at 100%. This data represents the results of one experiment with 3 mice/group at each of 4 timepoints. **P<0.01; ***P<0.001; ****P<0.0001 by two-tailed unpaired t-test.

Discussion

The goal of these studies was to assess the role of the stress-response gene REDD1 in the development of Oxa-induced ACD in mice. We demonstrate that REDD1 exerts striking and widespread pro-inflammatory effects in this disease model. REDD1 has a profound effect on both ACD phases: i) it supports the initial expansion of CD4⁺ and CD8⁺ cells and Th1/Th17/Th2 differentiation in draining lymph nodes (sensitization), and ii) it promotes the local edema response and infiltration of CD4⁺ and CD8⁺ T cells, macrophages, DCs and neutrophils into the ear after challenge (elicitation). In agreement with the reduced numbers of infiltrating leukocytes in challenged ears of Redd1 KO mice, the induction of many pro-inflammatory cytokines and chemokines was also significantly suppressed. The diminished early T cell response after sensitization in mice lacking REDD1 is the key deficit that sets the stage for the subsequent muted inflammatory responses observed in challenged ears. Lower numbers of activated T cells and a reduced capacity to produce cytokines results in fewer T cells available to migrate to the inflamed tissues to become reactivated. This is turn leads to less recruitment of other immune cells. However, there is also a clear migratory defect in REDD1-deficient T cells as shown by the reduced infiltration relative to WT animals. The diminished expression of Cxcr3 and Ccr4 in Redd1 KO mice, which regulate tissue migration of Th1 and Th2 cells respectively, supports this idea. It was previously shown that REDD1 is inducibly expressed in T cells in vitro (17), so it is likely that it is acting intrinsically to regulate activation, including expression of tissue homing molecules, in this disease setting. Yet this protein is also expressed in other cells and the possibility exists that APCs in the secondary lymphoid organs are impaired in T cell activating ability in KO mice. Studies are ongoing to distinguish between these scenarios.

There is a notably higher proportion of resident immune cells in the ear skin of naïve mice in Redd1 KO mice. The overall lack of response to challenge suggests that these cells are less responsive to activating signals as well and that REDD1 impacts the functionality of innate immune cells and perhaps regulates their homeostatic maintenance and migratory capacity.

Because REDD1 is best known as an mTOR inhibitor, we predicted that REDD1 deletion would relieve mTOR inhibition and promote inflammation, leading to increased T cell activation and immune cell trafficking. Indeed, there is evidence that mTOR activation, rather than inhibition, is implicated in T cell-mediated inflammation. Antigen receptor-mediated activation of resting T cells involves mTOR-dependent protein synthesis and an increase in glycolysis, necessary for rapid T cell proliferation and cytokine-driven differentiation and leading to variable effector functions and chemokine-induced migration to the site of inflammation. CD4⁺ T cells lacking mTOR fail to differentiate into Th1, Th2, or Th17 effector cells and instead, default to the Foxp3⁺ regulatory T cell differentiation pathway (24). Blockade of mTOR signaling in antigen-activated CD8⁺ T cells leads to the development of memory CD8⁺ T cells instead of effector cytotoxic T cells (24). mTOR also regulates T cell trafficking, promoting T cell egress from LNs (25). The suppressed T cell response to Oxa treatment in Redd1 KO mice we observe may reflect the differences of the controlled conditions of in vitro studies compared to the more complicated in vivo disease settings in which multiple cell types and signaling pathways are functioning. More likely, some pro-inflammatory effects of REDD1 do not require mTOR, an idea supported by several studies. For example, Redd1 KO mice exhibit decreased inflammation in an endotoxic shock model that is mTOR-independent (13). Based on decreased ROS production and Nox-1 and Gpx3 expression in these mice, it was proposed that REDD1 acts via the induction of oxidative stress. REDD1 is also known to activate the pro-inflammatory NF-κB pathway via an atypical mechanism mediated by physical interaction with its inhibitor, IkBα (9, 12). Finally, the REDD1-HIF-1α feedback loop that mediates adaptive responses to oxidative stress, important for some inflammatory processes, also does not involve mTOR (5). Our observation of reduced Nox1 gene expression in Redd1 KO mice supports a role for oxidative stress in promoting ACD. Reduced expression of inflammasome related genes, which are activated by ROS, including Nlrp3, Casp1 and Il1b is also consistent with this idea.

The actions of glucocorticoids are mediated by the glucocorticoid receptor (GR), a well-studied transcription factor (26). REDD1 is a known GR target gene, and as we reported previously, REDD1, via a feed-forward loop, controls GR signaling and overall transcriptional response to glucocorticoids, especially gene activation important for gluconeogenesis and atrophic adverse effects of glucocorticoids [16]). At the same time, we found that neither induction of inflammation nor glucocorticoid anti-inflammatory responses were affected in Redd1 KO animals treated with the skin irritant croton oil (CO) [16]. Unexpectedly, in Oxa-induced ACD, FA was less effective at inhibiting inflammation in Redd1 KO mice at lower doses (25 and 250 ng/animal), as measured by the reduced cellular infiltration to the ear and suppression of pro-inflammatory cytokines (Figure 8). There are major differences in the cellular mechanisms initiating inflammation in these two models. The major active component of CO is phorbol 12-myristate 13-acetate (PMA) [27], which directly induces keratinocytes to produce prostaglandins and IL-1β [12, 28] to initiate the inflammatory response [12, 29, 30]. In the ACD model, induction of inflammation is dependent on hapten-peptide complexes activating T cells, which triggers proliferation and cytokine production resulting in inflammation [14]. Using Illumina oligoarray, we previously found that REDD1 status did not significantly interfere with down-regulation of pro-proliferative and pro-inflammatory genes in epidermal keratinocytes in mouse skin treated with FA [16]. However, the changes in glucocorticoid transcriptome in Redd1 KO T cells remain to be investigated.

In summary, our data provides the first evidence that REDD1 is necessary for the full induction of inflammation associated with allergic contact hypersensitivity and affects both the sensitization and elicitation phase of the disease as well as the response to glucocorticoids. Although REDD1 is an inhibitor of mTOR, other mechanisms, including cell-specific mechanisms, may be involved in the pro-inflammatory role of this factor. The potent actions of REDD1 in inflammatory disease makes it an attractive target for therapeutics and provides strong rationale for further mechanistic studies.

Key points:

• REDD1 drives inflammation in oxalozone induced allergic contact dermatitis

• REDD1 is required for T cell activation and immune cell migration to affected tissues

• REDD1 may be attractive target for therapeutics in T cell mediated diseases

Abbreviations:

ACD

allergic contact dermatitis

CCR

chemokine C-C motif receptor

CO

croton oil

CXCL

chemokine C-X-C motif ligand

CXCR

chemokine C-X-C motif receptor

DC

dendritic cell

FA

fluocinolone acetonide

FACS

fluorescence-activated cell sorter

FMO

fluorescence minus one

H&E

hematoxylin and eosin

HIF-1α

hypoxia-inducible factor-1 α

IFNγ

interferon- γ

IL

interleukin

ILC2

type 2 innate lymphoid cells

LN

lymph node

LPS

lipopolysaccharide

mTOR

mammalian target of rapamycin

NF-κB

nuclear factor- γ B

PMA

phorbol 12-myristate 13-acetate

REDD1

regulated in development and DNA damage response

Oxa

oxazolone

ROS

reactive oxygen species

TNF

tumor necrosis factor

Th

T-helper cells

TSC

tuberous sclerosis complex

TSLP

thymic stromal lymphopoietin

WT

wild type