METHODS AND PHARMACEUTICAL COMPOSITION REDUCING SKIN INFLAMMATION

Abstract

The skin is one of first lines of defense against external threats. Tissue-resident macrophages have pivotal functions in tissue-barrier integrity and homeostasis. Upon skin inflammation, a functional crosstalk between the sensory nervous system and tissue-resident immune cells can regulate cutaneous immune responses. However, depending on the pathological context, sensory neurons display pro- or anti-inflammatory regulatory properties. Here the inventors identify, in a model of ultraviolet (UV)-induced skin N damage, a regulatory role for type C low-threshold mechanoreceptor (C-LTMR) sensory neurons on the dynamic of dermal macrophage replacement by inflammatory monocytes through the neuropeptide TAFA4. Tafa4-KO mice present an unresolved fibrotic dermis after UV irradiation. Increased fibrotic score correlates with the upstream persistency of inflammatory monocytes and their MHC-II.sup.+ macrophage progeny. Bone marrow chimera revealed that inflammatory monocyte differentiation towards CD206+ dermal macrophage is increased in Tafa4KO recipient. Finally, intradermal injection of TAFA4 at the site of UV irradiation reduces inflammatory monocytes accumulation and skin inflammation in Tafa4-KO mice. The results provide new insight about tissue-resident macrophages dynamic during the resolution of skin fibrosis and thus renders credible the use of TAFA4 for the treatment of skin inflammation.

Claims

1. A method of reducing skin inflammation in a subject in need thereof comprising administering to the subject a therapeutically effective amount of a TAFA4 polypeptide or a nucleic acid molecule encoding the TAFA4 polypeptide.

2. The method of claim 1 wherein the subject suffers from an inflammatory skin disease.

3. The method of claim 1 wherein the subject suffers from an inflammatory skin disease selected from the group consisting of acne, rosacea, folliculitis, perioral dermatitis, photodamage, skin aging, psoriasis, ichtiosis, chronic wounds, bed sores, keratosis piralis, scars, including surgical and acne scars, sebaceous cysts, inflammatory dermatoses, post inflammatory hyperpigmentation, xerosis, pruritis, lichen planus, nodular prurigo, eczema, and miliaria.

4. The method of claim 1 wherein the subject suffers from an inflammatory skin disease selected from the group consisting of scleroderma, atopic dermatitis, nephrogenic fibrosing dermopathy, mixed connective tissue disease, scleromyxedema, scleredema, keloid, sclerodactyly, and eosinophilic fasciitis.

5. The method of claim 1 wherein the subject suffers from photodermatitis.

6. The method of claim 1 wherein the subject suffers from a chronic wound.

7. The method of claim 1 wherein the subject suffers from venous stasis ulcers or diabetic foot ulcers.

8. The method of claim 6, wherein the chronic wound occurs in a subject suffering from sickle-cell disease or in an elderly subject.

9. (canceled)

10. The method of claim 1 wherein the TAFA4 polypeptide comprises a sequence having at least 90% sequence identity to the sequence as set forth in SEQ ID NO: 1.

11. The method of claim 1 wherein the TAFA4 polypeptide is fused to an immunoglobulin constant domain to constitute an immunoadhesin.

12. The method of claim 1 wherein the nucleic acid molecule is included in a vector.

13. The method of claim 1 wherein the TAFA4 polypeptide or the nucleic acid molecule encoding thereof are formulated for topical administration.

14. The method of claim 13 wherein the topical administration is performed via a transdermal device or a patch device.

15. The method of claim 1 wherein the TAFA4 polypeptide or the nucleic acid molecule encoding the TAFA4 polypeptide is administered with at least one other active agent.

16. The method of claim 12, wherein the vector is a viral vector.

17. The method of claim 16, wherein the viral vector is an adeno-associated virus (AAV) vector.

18. The method of claim 15, wherein the at least one other active agent is a glucocorticoid.

19. A method of preventing skin fibrosis in a subject in need thereof comprising administering to the subject a therapeutically effective amount of a TAFA4 polypeptide or a nucleic acid molecule encoding the TAFA4 polypeptide.

Description

FIGURES

[0037] FIG. 1: (A) WT (CTL), Tafa4-KO mice were treated with UV and ear thickness was followed over time. Tafa4-KO mice were treated either with saline (Tafa-4 KO or recombinant TAFA4 protein (100 nM) (Tafa-4 KO rescue). N=10 mice per group. (B-C) Absolute number of CCR2.sup.+ myeloid cell (B) or MHC-II.sup.+ macrophages (C) was assessed at day 7 post-UV treatment. All data are represented as mean+/−SEM. (*p<0.05; **p<0.01; ***p<0.001).

EXAMPLE

[0038] Material & Methods

[0039] Mice

[0040] Mice were maintained under standard housing conditions with free pathogen, free access to food and water and a 12 h light and dark cycle at an ambiant temperature of 22° C. in the animal facility of the CIML or CIPHE (CIML, France). C57/Bl6J mice were bought from Janvier (https://www.janvier-labs.com) and TAFA4-KO and GINIP-DTR mice were born in our animal house. TAFA-4-KO mice and NaV1.8CRE-GINIP-DTR mice were generated by Aziz Moqrich's team (IBDM, AMU, France) and described previously.sup.17,22. CCR2-KO mice were described previously.sup.31. Special effort was made to minimize number and stress. All protocols are in agreement with European Union recommendations for animal experimentation. All mice were used between 8 and 12 weeks unless specified.

[0041] Bone Marrow Chimera generation Age and sex matched WT or TAFA4KO mice were anaesthetized with ketamine/xylazine (10 μl/g, 2% Imalgene500 et 5% Rompun). Then mice were lethally irradiated with 6.5 Gy from an X-ray irradiator. A lead shield was used for protecting the ear's skin from the irradatiation-induced damage. 6 h after, 30 mg/kg busulfan was injected (i.p.) to deplete remaining myeloid progenitors in recipient mice. 12 h after these mice were reconstituted with 5-10 million bone marrow cells from CD45.1 or CCR2-KO mice. After irradiation, mice were treated with bactrim (in drinking water) for two weeks for a total recovery time of 5 weeks before analyzing the chimerisms efficiency and used for experiments.

[0042] Skin Injury

[0043] Age and sex matched WT, TAFA4KO or GINIP-DTR mice were anaesthetized with ketamine/xylazine (10 μl/g, 2% Imalgene500 et 5% Rompun) and then left untreated or exposed to UV (wavelength: 254 nm; voltage: 8 W; source: 30 cm from the target) for 30 min as previously described.sup.42. Ear thickness was followed every other day with a caliper (mice placed under isofluorane anesthesia) and assessed as thickness of UV-treated ears minus thickness of controls untreated ears.

[0044] Histology and Anatomopathology Analysis

[0045] At different time after UV treatment, mice were lethally anesthetized ketamine/xylazine (15 μl/g, 20% Imalgene500 et 5% Rompun) then 100 ml of blood was collected with glass Pasteur pipette. Mice were then perfused with 10 ml of PBS. Ears were then collected and fixed with formol (for at least 1 h then cryoprotected with sucrose 30% overnight and embedded in paraffin. Skin sections of 5 μm were stained with haematoxylin and eosin or picrosirius red. Histological scores were blinded assessed by a pathologist following the criteria described in table 2S. Briefly grading of inflammation, of wound healing and fibrosis with a semi-quantitative evaluation of fibrosis amount on one section and of the thickness of epidermis was made.

[0046] Dorsal Root Ganglia (DRG) Isolation and Tissue Section

[0047] After anesthesia (ketamine/xylazine mix in i.p.: 15 μl/g mice, 5% Rompun at 2%+20% Imalgen 500), mice were injected with 5 ml of cold PBS and right after with 25 ml of AntigenFix (PFA 4%) (i.c.). DRG were carefully dissected from the spinal cord under a binocular and collected in AntigenFix (PFA 4%). Then DRG were post-fixed in AntigenFix (PFA 4%) for at least 1 h then cryoprotected with 30% (w/v) sucrose overnight before being frozen in OCT and stored at −80° C. DRG samples were sectioned at 12 μm using a standard cryostat (Leica).

[0048] In Situ Hybridization and Immunofluorescence on DRG

[0049] In situ hybridization was carried out following protocol from.sup.16. Briefly, RNA probes were synthesized using gene-specific PCR primers and cDNA templates from mouse DRG.

[0050] DRG sections were treated with proteinase K then thriethanolamine and acetic anhydride solutions. Digoxigenin labeled probes were hybridized overnight at 55° C. after 2 h of prehybridization. The slides were treated with 0.2×SSC baths then blocked at room temperature with 10% Goat serum and incubated with anti-digoxigenin antibody (Roche). Final detection was achieved using Cyanine 3 TSA plus kit (Perkin Elmer).

[0051] For immunofluorescence, DRG sections were permeabilized with 0.3% Triton X-100 in FACS buffer for 1 h at room temperature. Primary antibodies were incubated as follows: rabbit anti-ATF3 1:200 (Santa Cruz SC-188), rat anti-TAFA4 (clone 1D8 1:500; MiMabs) overnight at 4° C., rat anti-Ginip 1:1000 (generous gift from Dr. A. Moqrich, IBDML) overnight at 4° C., goat anti-CGRP 1:1000 (Abcam Ab36000) 2 h at 37° C. Corresponding secondary antibodies (Alexa405, 488 or 594; Jackson ImmunoReasarch, 712-585-153, 712-475-153, 711-545-152, 705-585-147) were incubated 45 minutes at room temperature. Isolectin B4 conjugates with AlexaFluor 647 dye (Invitrogen) was used at 1:200, 45 minutes at room temperature.

[0052] Acquisition of images was performed on Confocal LSM780 (Zeiss) and analyzed with ZEN and ImageJ software.

[0053] Skin Cells Isolation

[0054] After UV treatment, mice were euthanized. Ears were split into dorsal and ventral layer then minced into small pieces and incubated 1 h at 37° C. in complete medium (RPMI+L-Glutamine 10% SVF) auditioned with 1 mg/ml DNase (Roche) 0.2 ml/ml Dispase (GIBCO) and 0.2 mg/ml collagenase type IV (Sigma). Then tissues were dissociated using 2.5 ml syringes and 18G needles and filtered on cell strainer (100 mm, BD), washed with Facs buffer (PBS-2 mM EDTA, FCS 2%, BSA 0.5%) to obtain a homogeneous cell suspension ready for staining.

[0055] Antibody Single-cell suspension were plated in 96 well U bottom plates and stained at 4° C. Dead cells were removed by counter-gating on the Livedead fixable blue Dead Cell Stain kit UV (L23105; Invitrogen). Antibodies used: CD45-BV785 (30F11; Biolegend), CD45.1-BV605 (A20; Biolegend), CX3CR1-BV421 (SA011F1), CD11b-BV510 (M1/70; BD Biosciences), CD64-BV711 (X54-5/7.1; Biolegend), CD169-BV605 (3D6.112; Biolegend), Ly6C-FITC (AL-21; BD Biosciences), F4-80-PECF594 (T45-2342; BD Biosciences), CD11c-PECy7 (N418; Biolegend), CCR2-PE (475301; R&D System), SiglecF-PE (E50-2440; BD Biosciences), CD206-APC (C068C2; Biolegend), Ly6G-APC-Cy7 (1A8; Biolegend), IA-IE-A700 (M5/114.15.2; Biolegend), CD24-BUV395 (M1/69; BD Biosciences), CD103-biotin (M290; BD Bioscience), MerTK-biotin (polyclonal, BAF591 R&D System), IL-1b-PE Ab (BD) anti-TNFa-PE (BD) Streptavidin-BUV737 (BD Biosciences).

[0056] Flow cytometry Cells were incubated 40 min at 4° C. in Facs buffer with antibody+2.4 G2 antibody to block Fc receptors, washed and fixed until analysis. For ex-vivo cytokines staining, cells suspensions were permeabilized using (fixperm FoxP3 kit; ebioscience) and stained with anti-IL-1b and TNF-α antibody. Multiparameter FACs analysis was performed using an LSR X20 system (BD). Absolute numbers for each population were obtained using Quanti Beads (556296, BD Bioscience). FACs analysis was performed using Flowjo software (Tree Star, Inc.).

[0057] Bone marrow derived macrophage generation and in vivo treatment Femurs and tibia bone marrow cells were flushed with sterile PBS, filtered (cell strainer, 100 mm, BD) and then cultured in complete DMEM (DMEM, SVF 10%, P/S, 1% L-Glu) supplemented with 10% L929 cell-conditioned medium. After 7 days in culture, mature BMDMs were harvested by washing with cold PBS, incubating on ice for 15 min, and pipetting extensively. 1×106 BMDMs/well were plated onto 12-well plates (BD) in complete DMEM medium supplemented with 100 ng.Math.mL M-CSF) and incubated with alternatively activated macrophage-inducing stimuli (10 ng.Math.mL IL-4 Peprotech) with or without 10 or 100 nM TAFA-4 (R&D System). After 16 hours, BMDM were lysed using RLT medium (Qiagen) then pass through a QIAShredder column (Qiagen) for homogenysation and store at −80° C.

[0058] Thyoglycolate Generated Macrophage and In Vitro Treatment

[0059] Peritoneal macrophages were isolated from C57Bl/6J (aged 8 to 12 weeks) from a peritoneal wash three days after 3% thyoglycolate injection i.p. (Sigma). Cells were seeded at 1×106 cells per well in complete DMEM (DMEM, FCS 10%, P/S, 1% L-glu) and incubated with or without LPS (100 ng.Math.ml, Sigma) in presence of increasing concentration of TAFA-4 (R&D System; 0, 1, 10, 100 or 1000 nM). After 8 hours, macrophages were lysed with a lysing medium (RLT, QIAGEN) then centrifuged in QIAShredder column for homogeinisation to preserve nucleic acid which were conserved at −80° C.

[0060] Gene Expression Analysis

[0061] Total RNA was isolated from ears untreated or treated with UV irradiation. Ears were dilacerated by FastPrep-24 (MpBio) with matrix A beads (MpBio). RNA was isolated using a fibrous RNeasy minikit (QIAGEN). Reverse transcription was performed using Superscript RTII (Invitrogen). Preamplification was performed using Taqman probe for gene target with pre-amplification master mix (Fluidigm). Pre-amplified products (18 cycles) were diluted five times before analysis with Universal PCR Master Mix and Taqman gene expression assays in 96.96 Dynamic Arrays on a BioMark System (Fluidigm). Data were analysed on the BioMark Real-Time PCR Analysis (Fluidigm) and normalised on a housekeeping gene, GADPH for in vitro experiment and HPRT for in vivo experiment (2{circumflex over ( )}-ΔCt). All Taqman gene were bought from Applied Biosystems and were the recommended one by the seller.

[0062] Statistical Analysis

[0063] Results are expressed as means+/−SEM. Statistical analysis was performed using the GraphPad Prism for Windows software. Statistical significance of the data was compared using the student's t test or Mann and Witney test if 2 group were compared. If multiple group, 2-way annova or multiple t test with bonferroni correction were used. Difference were considered significant as following: *p<0.05; **p<0.01; ***p<0.001; and ****p<0.0001.

[0064] Results and Discussion:

[0065] UV radiation causes sunburn-like damage characterized by the destruction of the epidermis and inflammation of the underneath dermal papilla.sup.8,9. These events lead to the rapid activation of mechanisms orchestrated by resident dermal macrophages and infiltrating monocytes to resolve inflammation and promote tissue repair.sup.10. Excessive inflammation can however lead to chronic tissue damage followed by excessive collagen deposition resulting in unresolved fibrotic scars. Tissue-specific signals constantly shape resident macrophage functional identity.sup.11,12 13 and promote their maintenance throughout the lifespan, either by local self-renewal or by the recruitment of additional monocyte-derived cells. However, the nature of these signals remains largely unknown. We hypothesized that the highly developed sensory neuron network present within the skin could provide key signals required for myeloid cell functions in tissue repair.

[0066] To understand how this potential neuro-immune crosstalk could occur after skin injury, we exposed the ears of wild type (WT) mice to UV-C irradiation to induce skin damage in sterile conditions (data not shown). The skin has a highly developed sensory nervous system that detects external stimuli and tissue lesions. Upon UV exposure, sunburned areas are generally characterized by a temporal phase of hypersensitivity to pain mediated by the activation of specialized sensory neurons innervating the skin.sup.14. We analyzed the kinetic of activation of cutaneous neurons after UV exposure. First, we traced sensory neurons innervating the ears by injecting the fluorescent dye DiI intradermally in the ear and analyzing the dorsal root ganglia (DRG) neurons for the presence of the dye within their cell body by confocal microscopy (data not shown). We analyzed the C2, C3, C4 DRGs and the trigeminal ganglia (TG) and found that the C2 and C3 were the DRG harboring the higher level of DiI staining showing that these DRG were the ones innervating the ears. The induction of the activating transcription factor 3 (ATF3) in DRG neurons was previously shown as a sensitive cellular marker to reveal the primary afferent fibers that are engaged by diverse chemical or noxious stimuli.sup.15. We investigated the capacity of skin innervating neurons to react to UV exposure by monitoring the expression profile of Atf3 by qRT-PCR in pooled C2/C3 DRGs over a time course of 14 days post-UV treatment (data not shown). We observed that, compared to non-exposed mice, UV-irradiation induced a significant expression of ATF3 mRNA in DRGs within one day. The maximal expression of Atf3 was reached by day 3 and maintained at a high level at least until day 7 post UV-irradiation. Residual expression of Atf3 gene was still observed at day 14 post-UV exposure. We then followed the kinetic of skin inflammation in our model by analyzing in the skin the expression profile of genes involved in myeloid cell activation, inflammation and tissue repair, at different time point post-irradiation by qRT-PCR (data not shown). This analysis revealed that inflammatory genes encoding factors such as IL1β, TNF-α, CXCL2, CCL2, CCL3, CCL6 and CXCL10 were up-regulated between day 3 and day 7 post-UV irradiation highlighting the inflammatory phase induced by UV. The expression of these genes slowly decreased afterwards, while the expression of genes involved in tissue repair such as Col1a and RETLNa raised only after day 14 until at least day 35, highlighting the phase of recovery and the onset of tissue remodeling. These data suggest that the kinetic of activation of sensory neurons within DRG correlates with the degree of inflammation observed within the skin exposed to UV (data not shown).

[0067] Upon skin invasion with pathogens or exposure to chemicals the deletion of subsets of skin sensory neurons can be either protective or deleterious for host defense and were found to induce either pro- or anti-inflammatory responses.sup.3 4 5 6. The variety of functional outcomes observed upon sensory neuron deficiencies.sup.2 could be explained by the complexity and diversity of sensory neurons subtypes.sup.7 bur their respective role in controlling immune cell functions and the molecular basis involved remain unclear.

[0068] To delineate the phenotype of the sensory neurons activated by UV irradiation, we performed immunofluorescent staining and confocal microcopy analysis of C3 DRGs of WT mice exposed or not to UV. At day 3 post-UV, ATF3 staining was observed in the nucleus of a subset of sensory neurons expressing the Gαi-Interacting Protein (GINIP) in exposed mice but not in control conditions (data not shown). Interestingly, this translocation of ATF3 was not observed in CGRP.sup.+ peptidergic neurons in UV-exposed mice (data not shown). These data suggest that GINIP.sup.+ neurons were preferentially activated by UV treatment. As their potential role on the control of cutaneous inflammation was not addressed so far, we decided to dissect their specific role in tissue repair in the context of UV irradiation-induced skin damage. We used a genetic model allowing the tissue specific and inducible ablation of GINIP-expressing neurons.sup.16. This model was obtained by crossing GINIP′ line.sup.17 with mice expressing the CRE recombinase from Nav1.8 locus (Nav1.8.sup.Cre/Cre mice).sup.18. Diphteria toxin (DT) treatment in Nav1.8.sup.Cre/+ GINIP.sup.flx/+ (hereafter called GINIP-DTR mice) at 4 weeks of age allowed the specific ablation of GINIP.sup.+ sensory neurons.sup.16. Nav1.8.sup.Cre/+ GINIP.sup.+/+ mice treated with DT were used as littermate controls throughout.

[0069] The ears of GINIP-DTR mice and littermate controls were exposed to UV irradiation 4 weeks after DT treatment. Tissue inflammation and damage was evaluated by measuring ear thickness every 2-3 days during 35 days (data not shown). We observed a strong inflammatory phase in both group between day 5 and day 10 post-UV exposition. Ear thickness was significantly increased in GINIP-DTR mice compared to control littermates starting at day 5 and until day 35 post-UV (data not shown). The peak of inflammation occurred at day 7 with a variation in ear thickness reaching 95 μm (+/−11 μm) in control mice versus 160 μm (+/−35 μm) in GINIP-DTR mice. The following recovery phase lasted 20 more days in the control group and more than 30 days in GINIP-DTR mice in which the ear thickness poorly reduced over time (data not shown). We then performed anatomo-pathology analysis at 14 and 35 days post-UV to better characterize the nature and extent of tissue damage in the two groups of mice (data not shown). Measuring the level of leukocyte infiltration (inflammation; data not shown), the thickening of the epidermis (data not shown) and collagen fiber expansion (fibrosis; data not shown), both groups reached a maximal score by day 14 suggesting massive tissue damages induced by UV-irradiation in both groups (data not shown). However, at day 35 leukocyte infiltration, epidermal thickness and fibrotic score were higher in GINIP-DTR mice compared to control littermates (data not shown). Accordingly, the cumulative score decreased significantly between day 14 and 35 in the control group but not in the GINIP-DTR mice group (data not shown), suggesting a defect in tissue remodeling in mice depleted for GINIP.sup.+ neurons. Further histological analysis of the ears at day 35 (data not shown), clearly showed a strong infiltration of collagen fibers throughout the whole dermis leading to the complete loss of the cartilage structure in ears of GINIP-DTR mice (data not shown) while the ears of control mice showed signs of fibrotic recovery (data not shown).

[0070] As tissue-resident macrophages and infiltrating monocytes are known to be central player in tissue remodeling and fibrosis, we analyzed the response of these cells in the skin of UV-exposed ears of GINIP-DTR and control littermates by flow cytometry. Excluding dendritic cells (CD11c.sup.+ MHC-II.sup.+), granulocytes (CD11b.sup.+ Ly6G.sup.+) and eosinophils (CD11b.sup.+ Siglec F.sup.+ CD24.sup.+) the monocyte/macrophage compartment, characterized by the double expression of CD11b and F480 (data not shown), can be further on separated on two subsets based on the expression of CCR2 and CD64 (data not shown). The CCR2.sup.+ population has been previously related to monocytes giving rise to monocytes-derived macrophages characterized by their high expression of MHC-II.sup.19 20. CD64 is commonly used to identify the tissue-resident macrophage populations.sup.21. Compared to control mice, GINIP-DTR mice exposed to UV irradiation presented a more persistent population of CCR2.sup.+ cells at day 14 (data not shown) as well as higher numbers of MHC-II.sup.+ macrophages between days 14 and 35 after UV irradiation (data not shown). Altogether, these data suggest that in the absence of GINIP.sup.+ neurons, the skin is more susceptible to UV-induced fibrosis and monocyte-derived cell infiltration.

[0071] GINIP is expressed in two distinct subsets of Ret.sup.+, non-peptidergic sensory neurons that can be distinguished through their differential staining with isolectin-B4 (IB4) and expression of TAFA4.sup.17. To delineate the underlying mechanisms leading to the strong skin damage observed in GINIP-DTR mice exposed to UV irradiation, we investigated by confocal microscopy on C3 DRGs which subsets of GINIP.sup.+ sensory neurons were reacting to UV exposure using a newly generated anti-TAFA4 monoclonal antibody, ATF3 and IB4 staining. The translocation of ATF3 was observed mostly in the nucleus of the IB4.sup.− TAFA4.sup.+ subset (data not shown), which represents C-LTMR.sup.17,22. TAFA4 molecule, a marker of this subset, is a chemokine-like protein.sup.23 known to modulates injury-induced mechanical and chemical pain hypersensitivity in mice.sup.22. Recombinant TAFA4 was shown to be involved in human macrophage chemotaxis in vitro.sup.24. However, its potential role on myeloid cell recruitment or activation in vivo is unknown. Therefore, we investigated if TAFA4 molecule could be involved in GINIP-DTR phenotype, by repeating UV-induced skin inflammation protocol on Tafa4-KO mice.sup.22. The monitoring of ears thickness in Tafa4-KO and control littermates, showed an inflammatory phase culminating for both control and Tafa4-KO mice between day 7 and day 10 post-UV treatment (data not shown). During the recovery phase that followed, the ear thickness was progressively resolved by day 35 in control mice. However, the ear thickness was significantly higher in Tafa-4KO compared to control mice from day 18 to 35 suggesting a defect in resolving skin inflammation and/or tissue remodeling in the absence of TAFA4.

[0072] To further characterize the phenotype of Tafa4-KO mice before and after UV-induced skin damage, we performed an anatomo-pathology analysis on their ear skin (data not shown). At steady state, without exposure to UV (data not shown), the skin of Tafa4-KO and littermate control mice were similar without any sign of inflammation. After UV-irradiation, the nature and extent of tissue damage were similar in the two groups of mice between day 7 and 21 post-UV (data not shown). In contrast, at day 35, as observed in GINIP-DTR mice, the scores measuring leukocytes infiltration (data not shown), epidermal thickness (data not shown) and collagen deposition (data not shown) were higher in Tafa4-KO mice than in littermates WT controls, with a cumulative score of 5.7+/−2.2 compared to 8+/−1.5 for control and Tafa4-KO mice respectively (data not shown). At this time point Tafa4-KO mice thus presented a persistent fibrosis of the dermis compared to the controls group. Consistent with this observation, Picro Sirius Red staining and histological analysis revealed excessive type 1 collagen deposition and persistence of unresolved fibrotic scars within the dermis of Tafa4-KO mice compared to littermate controls at day 35 (data not shown). Finally, flow cytometry analysis of skin cells in the ears of Tafa4-KO mice after UV showed, as seen in GINIP-DTR mice, the persistence of CCR2.sup.+ cells at day 14 (data not shown) and the presence of more abundant MHC-II.sup.+ macrophages at day 35 (data not shown). Therefore, Tafa4-KO mice phenocopy, at least partially, the fibrotic pathology and the changes in myeloid cell response observed in GINIP-DTR mice. Altogether these data suggest that the neuropeptide TAFA4 produced by a subset of GINIP.sup.+ neurons regulates myeloid cell responses to skin injury after UV irradiation.

[0073] Our data suggest a correlation between unresolved dermal fibrosis and the persistence of monocyte-derived cells within the skin of Tafa4-KO mice upon UV exposure. We further characterized the phenotype of macrophage subsets of the dermis using flow cytometry analysis (data not shown). As tissue repair is associated with functions of alternatively activated macrophage, we used CD206 marker to identify them in the dermis. We observed that most CCR2.sup.− CD64.sup.high macrophages were positive for this marker (data not shown) suggesting that CD206 also highlights tissue-resident macrophages of the dermis (data not shown). Analyzing the skin throughout the lifespan of WT mice using this gating strategy, we observed that CD206 was already expressed before birth, at E17.5 of embryonic development (data not shown) suggesting that CD206 marks long-lived, embryonic-derived tissue-resident macrophages of the dermis.sup.25 The CCR2.sup.+ population mostly consists in infiltrating monocytes (R1; Ly6C, MHC-II.sup.−), inflammatory monocytes (R2; Ly6C.sup.+ MHC-II.sup.+) and monocyte-derived macrophages (R3; Ly6C.sup.− MHC-II.sup.+) as observed here (data not shown) and in previous reports.sup.20. Indeed, this CCR2.sup.+ population is absent in embryonic dermis and only emerged within the first week after birth suggesting its relationship with bone marrow hematopoiesis (data not shown). Further analysis revealed that the CD206.sup.+ population contains two subsets of macrophages based on the expression or not of MHC-II (R4: MHC-II.sup.+ and R5: MHC-II.sup.−; data not shown). Using this gating strategy, we analyzed throughout the time course following UV exposure the dynamic of monocyte and macrophage subsets within the skin. UV treatment induced a massive recruitment of circulating monocytes, in both Tafa4-KO and littermate control mice, which culminated by day 5 and slowly decreased until day 23 (data not shown). No differences, in term of monocytes recruitment, were observed between the two groups. It is now well accepted that infiltrating monocytes progressively downregulate the expression of Ly6C and upregulate the expression of MHC-II to become inflammatory monocytes (data not shown). These inflammatory monocytes are known to produce IL-1-β and TNF-α as well as iNOS responsible for the production of reactive oxygen species (ROS) and can either give rise to Tip-dendritic cell or monocytes-derived macrophages depending on the tissue context.sup.26. Between day 14 to day 30 following UV irradiation, we observed an extended persistency of inflammatory monocytes within the dermis of Tafa4-KO mice compared to control littermates, suggesting that the absence of TAFA4 favors infiltrating monocyte to differentiate towards an inflammatory phenotype (data not shown). The monocyte-derived macrophage population (Ly6C.sup.− MHC-II.sup.+) also expanded significantly faster in Tafa4-KO than in control mice, at least until day 25 (data not shown). The resident dermal macrophages (R5), known to emerge before birth from embryonic precursors seeding the skin during organogenesis.sup.27, are long-lived cells maintained continually through local self-renewal without input from bone marrow circulating precursors.sup.28. However, several tissue-resident macrophage populations such as those of the gut, the heart as well as the dermis can still receive, although at a slow rate, some contribution from circulating monocytes to sustain their macrophage pool.sup.29 30 20. After UV treatment, the two subsets (MHC-II.sup.− and MHC-II.sup.+) of resident CD206+ dermal macrophages rapidly expanded until at least day 25 in both groups of mice (data not shown). Importantly, the number of dermal macrophages was higher in the skin of Tafa4-KO mice from day 7 to day 25. Combining the different macrophage populations identified in the dermis, the overall macrophage pool strongly expanded until day 35 and remarkably more rapidly in Tafa4-KO mice compared to control littermates (data not shown). This set of data first shows that UV irradiation triggers the recruitment of a massive number of monocytes from the circulation to provide additional macrophages locally probably supporting the dermal resident pool functions. These data show that TAFA4 is required for the control of dermal macrophage response triggered by UV exposure. Moreover, they highly suggest that the absence of TAFA4 leads to an excessive local expansion of dermal macrophage potentially responsible for the unresolved skin fibrosis phenotype.

[0074] As resident macrophages and monocyte-derived macrophages could have distinct function and contribution to the observed phenotype, we generated a series of bone marrow (BM) chimera in which the ears skin and their resident cells are protected from radiation by a lead shield (Ajami et al., 2007; data not shown). Monocytes circulation from the bone marrow to inflammed tissues highly relies on the CCR2/CCL2 axis.sup.31. To get insight on the monocyte contribution to the different macrophage subsets, we followed the expansion, in absolute number, of each population described earlier, 7 days post-UV, either in WT BM chimera mice (data not shown) or in WT chimera reconstituted with BM from CCR2-KO mice (data not shown). In CCR2-KO BM chimera mice, monocytes, inflammatory monocytes and monocytes-derived macrophages could not reach the inflammed dermis after UV treatment and could not contribute to the necessary expansion of these populations in the inflamed dermis (data not shown). As expected, the resident dermal macrophage population (data not shown) however could still expand independently from circulating precursors although with a noticed reduction compared to WT BM chimera mice. Interestingly, the MHCII+ CD206+ macrophage subset was rather highly dependent from circulating precursors. This suggests that this population could represent an intermediate subset bridging circulating monocytes and the long-lived bona fide resident dermal macrophage population (data not shown).

[0075] To further investigate how the microenvironment impacts monocytes fate in the absence of TAFA4, we reconstituted CD45.2 WT or Tafa4-KO irradiated mice with CD45.1 BM. In these conditions infiltrating monocyte fate can be traced based on CD45.1 expression while resident dermal macrophages (R5) could be identified through CD45.2 expression. Resident macrophages expanded drastically after UV irradiation and even faster in Tafa4-KO mice compared to controls littermates (data not shown) in both WT and Tafa4-KO chimera mice unexposed to UV, five weeks after reconstitution, the contribution of WT CD45.1 BM to the resident dermal macrophage pool (R5) remained below 25% (data not shown). As expected, the frequency of CD45.1.sup.+ brain microglia, which remains independent from circulating monocytes reached less than 3%, while circulating monocytes and granulocytes reached 95% (data not shown). These data suggest a slow but consistent turnover of dermal-resident macrophages. In contrast, upon exposure to UV, the contribution of CD45.1+BM circulating precursors to dermal macrophages reached 55% (+/−8%) in WT versus 74% (+/−9%) Tafa4-KO recipient mice, 14 days after UV treatment. Two alternative explanations could underlie these data. Resident dermal macrophages could be UV sensitive, leading, after their elimination for example through necroptosis mechanisms.sup.32 to their replacement by circulating monocytes. Alternatively, the resident dermal macrophage pool could remain stable but their relative frequency would be drowned following a massive expansion of monocyte-derived cells which can differentiate into CD206.sup.+ MHCII.sup.+ intermediate subset as observed earlier (data not shown, UV groups). The analysis of the same cell subsets at day 35 post-UV treatment supports the second hypothesis as the chimerism of dermal macrophages returned to 41%+/−12% in WT recipient and 55%+/−13% in Tafa4-KO recipient (data not shown). Three months after UV exposure, the chimerism of dermal resident macrophages decreased to 34%+/−7% in WT recipient while it increased to 63%+/−12% in Tafa4-KO recipient mice, respectively (data not shown). These data suggest that monocytes differentiation towards resident dermal macrophages, observed in WT recipient mice is transient and that bona fide dermal macrophage expand during the recovery phase to take over their initial niche. In addition, our results suggest that in the absence of TAFA4, monocyte-derived dermal macrophages remain longer into the dermis and potentially become long-lived resident cells. The origin of dermal macrophages in the Tafa4-KO versus WT mice are thus different suggesting potential different functionality.

[0076] To further understand the mechanism by which TAFA4 could regulate myeloid cell response, we tested if recombinant TAFA4 was able to directly modulate their function in vitro. In contrast to a previous report.sup.24, we were not able to detect any chemotactic activity of TAFA4 on mouse macrophages (data not shown). However, we found that the transcriptomic profile of IL-4 primed BM-derived macrophages was modified in the presence of TAFA4. The expression of transcription factors encoding genes, such as IRF5 and IRF8, involved in developmental maturation of macrophages and polarisation.sup.33,34 were upregulated by increasing concentration of TAFA4 (data not shown). Moreover, the gene Nr4a1, required for the regulation of macrophage activation and apoptosis as well as for monocytes differentiation and Ly6C° monocyte survival.sup.35-37 was also upregulated by TAFA4 (data not shown). The mRNA encoding TGF-β was also upregulated in response to TAFA4 (data not shown). TGF-β, besides being a central factor for the resolution of inflammation, has been recently implicated in the maintenance of several resident macrophage as well as in monocyte recruitment and differentiation.sup.38,39. These data suggest that TAFA4 can directly act on macrophages and activate regulatory pathways involved in key functions of macrophages and monocytes. Finally, thioglycolate-induced inflammatory macrophages, treated in vitro with lipopolysaccharide.sup.40 downregulated, in response to additional TAFA4, the Receptor-interacting protein kinase 3 (RIPK3), a gene involved in necroptosis suggesting that TAFA4 could protect macrophages from death induced by excessive inflammation.sup.41 (data not shown).

[0077] These in vitro data suggest that TAFA4 could directly modulate myeloid cell homeostasis. Its overall absence in Tafa4-KO mice could favor tissue-resident macrophage replacement under inflammatory condition. Consistent with a potential role of TAFA4 in inducing an anti-inflammatory transcriptomic profile in macrophages, we found more Ly6C.sup.+ MHCII.sup.+ inflammatory monocytes producing IL-1-β and TNF-α in Tafa4-KO mice compared to littermate controls in vivo 7 days after UV exposure (data not shown). We showed that inflammatory cytokines and chemokines production, such as TFN-α, IL-1b, IL-6, CXCL1, CCL4 or CCL2 production, are extended in TAFA4-KO mice in total skin after 7 or 10 days of UV irradiation (Data not shown). The expression of IL-10 is upregulated in purified macrophage in WT mice after UV irradiation, but not in TAFA4 KO-mice. An ex vivo analysis of IL-10 production in macrophage subsets (Tim4+ Mac, DN Mac and MCHII+ Mac) was proceed. We found that two subsets of dermal macrophages, Tim4+ Mac and DN mac, upregulate IL-10 production in vivo following UV irradiation of the skin in WT mice but not in TAFA4 KO mice (data not shown). This suggests that TAFA4 could not only regulate resident cell biology but could also inhibit inflammatory functions of newly recruited monocytes into tissues.

[0078] To confirm the anti-inflammatory potential of TAFA4, we performed intradermal injections of TAFA4 (100 nM, 50 ul) or saline in the ears of Tafa4-KO mice every 2 days starting the day of UV treatment during 8 days. The follow up of ear thickness over time revealed a reduction of the inflammatory phase in the Tafa4-KO mice treated with TAFA4 (110 μm+/−11 μm) compared to Tafa4-KO or control mice treated with saline (145 uM+/−23 μm and 135+/−27 μm respectively) at day 10 (FIG. 1A). The reduction of ear thickness in TAFA4-treated versus saline-injected Tafa4-KO mice was also significant between day 14 and 19 post-UV exposure (FIG. 1A). However, at later time points, the ear thickness returned to the level observed in the Tafa4-KO control group treated with saline suggesting that the presence of TAFA4 is required during the whole process of tissue repair. We then analyzed if the reduction in ear thickness of Tafa4-KO mice treated with TAFA4 protein was associated with modifications in myeloid cell responses. Consistent with the phenotype described earlier, in mice injected with saline, a clear increase in the number of CCR2.sup.+ macrophages was observed in the ears of Tafa4-KO compared to control littermates, 7 days after UV irradiation. In contrast, Tafa4-KO mice treated with TAFA4 showed a rescue of this phenotype with CCR2+ macrophages in the ear skin (FIG. 1B). In addition, the local injection of TAFA4 also reduced the number of MHC-II.sup.+ macrophages in the ears of Tafa4-KO mice 7 days after UV irradiation (FIG. 1C). Therefore, injection of TAFA4 in injured skin can restore the impaired response observed in Tafa4-KO mice in vivo suggesting that such treatment may have a potential interest in some pathological conditions associated with skin inflammatory diseases.

[0079] Altogether, these data show that TAFA4, a molecule expressed by a subset of GINIP.sup.+ sensory neurons innervating the skin is required for the resolution of skin inflammation and fibrosis upon exposure to UV irradiation. Our results also suggest that treatment with TAFA4 in vitro and in vivo can promote an anti-inflammatory and pro-repair myeloid cell responses.

[0080] Primary sensory neurons are heterogeneous by numerous molecular criteria. The variety of functional outcomes observed upon sensory neuron deficiencies could be explained by the complexity and diversity of sensory neurons subtypes. However the functional significance of this remarkable heterogeneity is just emerging. Previous studies have shown that sensory neurons involved in pain sensitivity can modulate some aspects of skin inflammatory immune responses.sup.2. However, most of them focused on the role of peptidergic subsets and in particular on the role of the neuropeptide CGRP that has been described to have mainly pro-inflammatory effects.sup.3,4. Here we describe that a group of non peptidergic neurons expressing the marker GINIP are required to down-regulate inflammatory processes and to modulate the myeloid cell response. This control is required to promote tissue repair in the context of skin exposure to UV irradiation. We identified TAFA4 a factor produced by GINIP.sup.+ C-LTMR has a major factor involved in this regulation. However, the skin inflammatory phenotype in Tafa4-KO mice is less pronounced than in GINIP-DTR mice (data not shown) suggesting that other mediators expressed by GINIP.sup.+ neurons may have additional regulatory functions. This study reveals new mechanisms of neuro-immune regulations and paves the way for future investigations analyzing the role of this novel regulatory pathway in other skin inflammatory diseases.

REFERENCES

[0081] Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure. [0082] 1 Guilliams, M. & Scott, C. L. Does niche competition determine the origin of tissue-resident macrophages? Nat Rev Immunol 17, 451-460, doi:10.1038/nri.2017.42 (2017). [0083] 2 Pinho-Ribeiro, F. A., Verri, W. A., Jr. & Chiu, I. M. Nociceptor Sensory Neuron-Immune Interactions in Pain and Inflammation. Trends Immunol 38, 5-19, doi:10.1016/j.it.2016.10.001 (2017). [0084] 3 Kashem, S. W. et al. Nociceptive Sensory Fibers Drive Interleukin-23 Production from CD301b+ Dermal Dendritic Cells and Drive Protective Cutaneous Immunity. Immunity 43, 515-526, doi:10.1016/j.immuni.2015.08.016 (2015). [0085] 4 Riol-Blanco, L. et al. Nociceptive sensory neurons drive interleukin-23-mediated psoriasiform skin inflammation. Nature 510, 157-161, doi:10.1038/nature13199 (2014). [0086] 5 Chiu, I. M. et al. Bacteria activate sensory neurons that modulate pain and inflammation. Nature 501, 52-57, doi:10.1038/nature12479 (2013). [0087] 6 Pinho-Ribeiro, F. A. et al. Blocking Neuronal Signaling to Immune Cells Treats Streptococcal Invasive Infection. Cell 173, 1083-1097 e1022, doi:10.1016/j.cell.2018.04.006 (2018). [0088] 7 Usoskin, D. et al. Unbiased classification of sensory neuron types by large-scale single-cell RNA sequencing. Nat Neurosci 18, 145-153, doi:10.1038/nn.3881 (2015). [0089] 8 Merad, M. et al. Langerhans cells renew in the skin throughout life under steady-state conditions. Nat Immunol 3, 1135-1141. (2002). [0090] 9 Ajami, B., Bennett, J. L., Krieger, C., Tetzlaff, W. & Rossi, F. M. Local self-renewal can sustain CNS microglia maintenance and function throughout adult life. Nat Neurosci 10, 1538-1543, doi:nn2014 [pii] 10.1038/nn2014 (2007). [0091] 10 Wynn, T. A. & Vannella, K. M. Macrophages in Tissue Repair, Regeneration, and Fibrosis. Immunity 44, 450-462, doi:10.1016/j.immuni.2016.02.015 (2016). [0092] 11 Lavin, Y. et al. Tissue-resident macrophage enhancer landscapes are shaped by the local microenvironment. Cell 159, 1312-1326, doi:10.1016/j.cell.2014.11.018 (2014). [0093] 12 Gosselin, D. et al. Environment drives selection and function of enhancers controlling tissue-specific macrophage identities. Cell 159, 1327-1340, doi:10.1016/j.cell.2014.11.023 (2014). [0094] 13 Okabe, Y. & Medzhitov, R. How the immune system spots tumors. Elife 3, e04476, doi:10.7554/eLife.04476 (2014). [0095] 14 Lopes, D. M. & McMahon, S. B. Ultraviolet Radiation on the Skin: A Painful Experience? CNS Neurosci Ther 22, 118-126, doi:10.1111/cns.12444 (2016). [0096] 15 Braz, J. M. & Basbaum, A. I. Differential ATF3 expression in dorsal root ganglion neurons reveals the profile of primary afferents engaged by diverse noxious chemical stimuli. Pain 150, 290-301, doi:10.1016/j.pain.2010.05.005 (2010). [0097] 16 Urien, L. et al. Genetic ablation of GINIP-expressing primary sensory neurons strongly impairs Formalin-evoked pain. Sci Rep 7, 43493, doi:10.1038/srep43493 (2017). [0098] 17 Gaillard, S. et al. GINIP, a Galphai-interacting protein, functions as a key modulator of peripheral GABAB receptor-mediated analgesia. Neuron 84, 123-136, doi:10.1016/j.neuron.2014.08.056 (2014). [0099] 18 Abrahamsen, B. et al. The cell and molecular basis of mechanical, cold, and inflammatory pain. Science 321, 702-705, doi:10.1126/science.1156916 (2008). [0100] 19 Jakubzick, C. et al. Minimal differentiation of classical monocytes as they survey steady-state tissues and transport antigen to lymph nodes. Immunity 39, 599-610, doi:10.1016/j.immuni.2013.08.007 (2013). [0101] 20 Tamoutounour, S. et al. Origins and functional specialization of macrophages and of conventional and monocyte-derived dendritic cells in mouse skin. Immunity 39, 925-938, doi:10.1016/j.immuni.2013.10.004 (2013). [0102] 21 Gautier, E. L. et al. Gene-expression profiles and transcriptional regulatory pathways that underlie the identity and diversity of mouse tissue macrophages. Nat Immunol 13, 1118-1128, doi:10.1038/ni.2419 (2012). [0103] 22 Delfini, M. C. et al. TAFA4, a chemokine-like protein, modulates injury-induced mechanical and chemical pain hypersensitivity in mice. Cell Rep 5, 378-388, doi:10.1016/j.celrep.2013.09.013 (2013). [0104] 23 Tom Tang, Y. et al. TAFA: a novel secreted family with conserved cysteine residues and restricted expression in the brain. Genomics 83, 727-734, doi:10.1016/j.ygeno.2003.10.006 (2004). [0105] 24 Wang, W. et al. FAM19A4 is a novel cytokine ligand of formyl peptide receptor 1 (FPR1) and is able to promote the migration and phagocytosis of macrophages. Cellular & molecular immunology 12, 615-624, doi:10.1038/cmi.2014.61 (2015). [0106] 25 Hoeffel, G. et al. C-Myb(+) erythro-myeloid progenitor-derived fetal monocytes give rise to adult tissue-resident macrophages. Immunity 42, 665-678, doi:10.1016/j.immuni.2015.03.011 (2015). [0107] 26 Serbina, N. V., Salazar-Mather, T. P., Biron, C. A., Kuziel, W. A. & Pamer, E. G. TNF/iNOS-producing dendritic cells mediate innate immune defense against bacterial infection. Immunity 19, 59-70 (2003). [0108] 27 Hoeffel, G. & Ginhoux, F. Fetal monocytes and the origins of tissue-resident macrophages. Cell Immunol, doi: 10.1016/j.cellimm.2018 0.01.001 (2018). [0109] 28 Ginhoux, F. & Jung, S. Monocytes and macrophages: developmental pathways and tissue homeostasis. Nat Rev Immunol 14, 392-404, doi:10.1038/nri3671 (2014). [0110] 29 Bain, C. C. et al. Constant replenishment from circulating monocytes maintains the macrophage pool in the intestine of adult mice. Nat Immunol 15, 929-937, doi:10.1038/ni.2967 (2014). [0111] 30 Molawi, K. et al. Progressive replacement of embryo-derived cardiac macrophages with age. J Exp Med 211, 2151-2158, doi:10.1084/jem.20140639 (2014). [0112] 31 Serbina, N. V. & Pamer, E. G. Monocyte emigration from bone marrow during bacterial infection requires signals mediated by chemokine receptor CCR2. Nat Immunol 7, 311-317, doi:10.1038/ni1309 (2006). [0113] 32 Bleriot, C. et al. Liver-resident macrophage necroptosis orchestrates type 1 microbicidal inflammation and type-2-mediated tissue repair during bacterial infection. Immunity 42, 145-158, doi:10.1016/j.immuni.2014.12.020 (2015). [0114] 33 Lawrence, T. & Natoli, G. Transcriptional regulation of macrophage polarization: enabling diversity with identity. Nat Rev Immunol 11, 750-761, doi:10.1038/nri3088 (2011). [0115] 34 Hagemeyer, N. et al. Transcriptome-based profiling of yolk sac-derived macrophages reveals a role for Irf8 in macrophage maturation. EMBO J 35, 1730-1744, doi:10.15252/embj.201693801 (2016). [0116] 35 Palumbo-Zerr, K. et al. Orphan nuclear receptor NR4A1 regulates transforming growth factor-beta signaling and fibrosis. Nat Med 21, 150-158, doi:10.1038/nm.3777 (2015). [0117] 36 Hanna, R. N. et al. The transcription factor NR4A1 (Nur77) controls bone marrow differentiation and the survival of Ly6C− monocytes. Nat Immunol 12, 778-785, doi:10.1038/ni.2063 (2011). [0118] 37 Thomas, G. D. et al. Deleting an Nr4a1 Super-Enhancer Subdomain Ablates Ly6Clow Monocytes while Preserving Macrophage Gene Function. Immunity 45, 975-987, doi:10.1016/j.immuni.2016.10.011 (2016). [0119] 38 Yu, X. et al. The Cytokine TGF-beta Promotes the Development and Homeostasis of Alveolar Macrophages. Immunity 47, 903-912 e904, doi:10.1016/j.immuni.2017.10.007 (2017). [0120] 39 Lund, H. et al. Fatal demyelinating disease is induced by monocyte-derived macrophages in the absence of TGF-beta signaling. Nat Immunol 19, 1-7, doi:10.1038/s41590-018-0091-5 (2018). [0121] 40 Bogunovic, M. et al. Identification of a radio-resistant and cycling dermal dendritic cell population in mice and men. J Exp Med 203, 2627-2638, doi:10.1084/jem.20060667 (2006). [0122] 41 Vanden Berghe, T., Linkermann, A., Jouan-Lanhouet, S., Walczak, H. & Vandenabeele, P. Regulated necrosis: the expanding network of non-apoptotic cell death pathways. Nat Rev Mol Cell Biol 15, 135-147, doi:10.1038/nrm3737 (2014). [0123] 42 Greter, M. et al. Stroma-derived interleukin-34 controls the development and maintenance of langerhans cells and the maintenance of microglia. Immunity 37, 1050-1060, doi:10.1016/j.immuni.2012.11.001 (2012).