Advertisement

TNF-α and IL-1β Do Not Induce Langerhans Cell Migration by Inhibiting TGFβ Activation

  • Jacinto S. De La Cruz Diaz
    Affiliations
    Department of Dermatology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA

    Department of Immunology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
    Search for articles by this author
  • Toshiro Hirai
    Affiliations
    Department of Dermatology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA

    Department of Immunology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
    Search for articles by this author
  • Breanna Anh-Thu Nguyen
    Affiliations
    Department of Dermatology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA

    Department of Immunology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
    Search for articles by this author
  • Yukari Zenke
    Affiliations
    Department of Dermatology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA

    Department of Immunology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA

    Department of Dermatology, St. Luke’s International Hospital, Tokyo, Japan
    Search for articles by this author
  • Yi Yang
    Affiliations
    Department of Dermatology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA

    Department of Immunology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA

    Department of Dermatology, Xiangya Hospital, Central South University, Changsha, China

    Department of Dermatology, Xiangya Hospital, Central South University, Changsha, Hunan, China

    The Third Xiangya Hospital, Central South University, Changsha, Hunan, China
    Search for articles by this author
  • Haiyue Li
    Affiliations
    Department of Dermatology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA

    Department of Immunology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA

    School of Medicine, Tsinghua University, Beijing, China
    Search for articles by this author
  • Stephen Nishimura
    Affiliations
    Department of Pathology, University of California San Francisco, San Francisco, California, USA
    Search for articles by this author
  • Daniel H. Kaplan
    Correspondence
    Correspondence: Daniel H. Kaplan, Department of Dermatology, University of Pittsburgh, 200 Lothrop Street, BST W1046, Pittsburgh, Pennsylvania 15261, USA.
    Affiliations
    Department of Dermatology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA

    Department of Immunology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
    Search for articles by this author
Open AccessPublished:May 22, 2021DOI:https://doi.org/10.1016/j.xjidi.2021.100028
      In the skin, Langerhans cells (LCs) require autocrine latent TGFβ that is transactivated by the integrins ανβ6 and ανβ8 expressed by keratinocytes (KCs) for long-term epidermal retention. Selective expression of a ligand-independent, constitutively active form of TGFβR1 inhibits LC migration during homeostasis and in response to UVB exposure. In this study, we found that LC migration in response to inflammatory stimuli was also inhibited by ligand-independent TGFβR1 signaling. Contrary to UVB stimulation, which reduced KC expression of ανβ6, in vitro and in vivo exposure to TNF-α or IL-1β increased ανβ6 transcript and protein expression by KCs. This resulted in increased KC-mediated transactivation of latent TGFβ. Expression of ανβ8 was largely unchanged. These findings show that ligand-independent TGFβR1 signaling in LCs can overcome inflammatory migration stimuli, but reduced KC-mediated transactivation of latent TGFβ by KCs may only drive LC migration during homeostasis and in response to UV stimulation.

      Abbreviations:

      DMBA (7,12-dimethylbenz[a]anthracene), EpCAM (epithelial cell adhesion molecule), IFE (interfollicular), IM (infundibulum/isthmus), KC (keratinocyte), LAP (latency associated peptide), LC (Langerhans cell), LN (lymph node), MHC (major histocompatibility complex), pKC (primary keratinocyte)

      Introduction

      The skin serves as a physical barrier against a plethora of biologic and nonbiologic agents alike, including UV, chemical haptens, and commensal as well as pathogenic microbes. Keratinocytes (KCs) in the epidermis create a physical barrier but also form a niche for the cells of the immune system such as Langerhans cells (LCs), CD8+ tissue-resident memory T cells, and dendritic epidermal gamma delta T cells. These leukocytes, along with KC, work in harmony to maintain barrier integrity and orchestrate effective immune responses (
      • Kobayashi T.
      • Naik S.
      • Nagao K.
      Choreographing immunity in the skin epithelial barrier.
      ). Within the epidermis, LCs possess the unique ability to migrate from the skin epidermis to regional lymph nodes (LNs). At steady state, LCs migrate to LN transporting self and potentially commensal-derived antigen to promote the development of peripheral tolerance (
      • Hemmi H.
      • Yoshino M.
      • Yamazaki H.
      • Naito M.
      • Iyoda T.
      • Omatsu Y.
      • et al.
      Skin antigens in the steady state are trafficked to regional lymph nodes by transforming growth factor-beta1-dependent cells.
      ). During infection, LCs transport potential pathogen-derived antigen and prime naïve adaptive immune responses (
      • Igyártó B.Z.
      • Haley K.
      • Ortner D.
      • Bobr A.
      • Gerami-Nejad M.
      • Edelson B.T.
      • et al.
      Skin-resident murine dendritic cell subsets promote distinct and opposing antigen-specific T helper cell responses.
      ;
      • Kobayashi T.
      • Glatz M.
      • Horiuchi K.
      • Kawasaki H.
      • Akiyama H.
      • Kaplan D.H.
      • et al.
      Dysbiosis and Staphylococcus aureus colonization drives inflammation in atopic dermatitis.
      ).
      Inflammatory cytokines, including TNF-α and IL-1β, are well-known to trigger LC migration and are thought to directly interact with LCs (
      • Cumberbatch M.
      • Dearman R.J.
      • Kimber I.
      Langerhans cells require signals from both tumour necrosis factor-alpha and interleukin-1 beta for migration.
      ;
      • Cumberbatch M.
      • Fielding I.
      • Kimber I.
      Modulation of epidermal Langerhans' cell frequency by tumour necrosis factor-alpha.
      ). A direct effect on LCs is unlikely, at least for IL-1β, because LCs lacking Myd88, which is required for signaling through toll-like receptor and IL-1 receptor family members, migrate normally in response to IL-1β injection (
      • Didovic S.
      • Opitz F.V.
      • Holzmann B.
      • Förster I.
      • Weighardt H.
      Requirement of MyD88 signaling in keratinocytes for Langerhans cell migration and initiation of atopic dermatitis-like symptoms in mice.
      ;
      • Haley K.
      • Igyártó B.Z.
      • Ortner D.
      • Bobr A.
      • Kashem S.
      • Schenten D.
      • et al.
      Langerhans cells require MyD88-dependent signals for Candida albicans response but not for contact hypersensitivity or migration.
      ). LC migration is also unaffected in response to Candida albicans infection and DNFB application and during homeostasis. Another key cytokine involved in LC migration is TGFβ1 (TGFβ). LC-specific genetic ablation of Tgfb1, Tgfbr1 (ALK5), or Tgfbr2 results in fully differentiated LCs that spontaneously migrate from the epidermis into regional LNs (
      • Borkowski T.A.
      • Letterio J.J.
      • Farr A.G.
      • Udey M.C.
      A role for endogenous transforming growth factor beta 1 in Langerhans cell biology: the skin of transforming growth factor beta 1 null mice is devoid of epidermal Langerhans cells.
      ;
      • Kaplan D.H.
      Ontogeny and function of murine epidermal Langerhans cells.
      ;
      • Kaplan D.H.
      • Li M.O.
      • Jenison M.C.
      • Shlomchik W.D.
      • Flavell R.A.
      • Shlomchik M.J.
      Autocrine/paracrine TGFbeta1 is required for the development of epidermal Langerhans cells.
      ;
      • Kel J.M.
      • Girard-Madoux M.J.
      • Reizis B.
      • Clausen B.E.
      TGF-beta is required to maintain the pool of immature Langerhans cells in the epidermis.
      ). On synthesis and secretion, TGFβ is bound to the latency associated peptide (LAP), resulting in inactive TGFβ (i.e., LAP–TGFβ1) (
      • Travis M.A.
      • Sheppard D.
      TGF-β activation and function in immunity.
      ). Removal of the LAP protein from TGFβ1 can be accomplished by low pH, proteases, mechanical stress, ROS, and integrin-mediated processes (
      • Worthington J.J.
      • Klementowicz J.E.
      • Travis M.A.
      TGFβ: a sleeping giant awoken by integrins.
      ). In the epidermis, activation of LAP–TGFβ1 is accomplished through functionally active integrins ανβ6 and ανβ8 that are expressed on KC (
      • Aluwihare P.
      • Mu Z.
      • Zhao Z.
      • Yu D.
      • Weinreb P.H.
      • Horan G.S.
      • et al.
      Mice that lack activity of alphavbeta6- and alphavbeta8-integrins reproduce the abnormalities of Tgfb1- and Tgfb3-null mice.
      ;
      • Yang Z.
      • Mu Z.
      • Dabovic B.
      • Jurukovski V.
      • Yu D.
      • Sung J.
      • et al.
      Absence of integrin-mediated TGFbeta1 activation in vivo recapitulates the phenotype of TGFbeta1-null mice.
      ). Integrin activity is positively regulated through the cytoplasmic adaptor proteins Talin1 (Tln1) and Kindlin1 (Fermt1) (
      • Kerr B.A.
      • Byzova T.V.
      Integrin alpha V (ITGAV).
      ;
      • Qin J.
      • Vinogradova O.
      • Plow E.F.
      Integrin bidirectional signaling: a molecular view.
      ;
      • Takada Y.
      • Ye X.
      • Simon S.
      The integrins.
      ).
      Epidermal KCs can be categorized on the basis of their relative spatial relationship to the hair follicles as interfollicular (IFE), infundibulum/isthmus (IM), and bulge KCs (
      • Nagao K.
      • Kobayashi T.
      • Moro K.
      • Ohyama M.
      • Adachi T.
      • Kitashima D.Y.
      • et al.
      Stress-induced production of chemokines by hair follicles regulates the trafficking of dendritic cells in skin.
      ). At a steady state, LCs reside intercalated with IFE and IM KCs but are actively excluded from the bulge (
      • Nagao K.
      • Kobayashi T.
      • Moro K.
      • Ohyama M.
      • Adachi T.
      • Kitashima D.Y.
      • et al.
      Stress-induced production of chemokines by hair follicles regulates the trafficking of dendritic cells in skin.
      ). IFE KCs express integrin ανβ6 but not ανβ8, and IM KCs express ανβ8 but not ανβ6 (
      • Mohammed J.
      • Beura L.K.
      • Bobr A.
      • Astry B.
      • Chicoine B.
      • Kashem S.W.
      • et al.
      Stromal cells control the epithelial residence of DCs and memory T cells by regulated activation of TGF-β.
      ). Integrin β6 (Itgb6) and β8 (Itgb8) form obligate heterodimers with integrin αv (Itgav) to form functional integrin proteins (
      • Kerr B.A.
      • Byzova T.V.
      Integrin alpha V (ITGAV).
      ,
      • Takada Y.
      • Ye X.
      • Simon S.
      The integrins.
      ). Genetic ablation of Itgb6 results in loss of LCs from the IFE owing to increased spontaneous LC migration into regional LNs. Similarly, ablation of Itgb8 results in loss of LCs from the IM. Notably, LC-specific expression of a constitutively active, ligand-independent form of the TGFβR1 (TGFβRCA) prevents spontaneous LC migration in Itgb6–/– mice. Epidermal exposure to UVB is sufficient to reduce KC expression integrin ανβ6 or ανβ8 and induce efficient LC migration that can be inhibited by LC-specific expression of TGFβRCA (
      • Mohammed J.
      • Beura L.K.
      • Bobr A.
      • Astry B.
      • Chicoine B.
      • Kashem S.W.
      • et al.
      Stromal cells control the epithelial residence of DCs and memory T cells by regulated activation of TGF-β.
      ). Taken together, these data support a model where autocrine LC-derived LAP–TGFβ is transactivated either by ανβ6 or ανβ8 expressed by KCs and then acts directly on LCs to prevent spontaneous migration. Moreover, the data suggest that reduced TGFβ transactivation after UVB irradiation is a trigger for LC migration. Whether a similar mechanism occurs with inflammatory stimuli remains unexplored.
      In this study, we report that enforced TGFβ receptor signaling in LCs is sufficient to prevent inflammation-induced LC migration. UVB efficiently reduced the expression of ανβ6 and ανβ8 on primary KCs (pKCs), resulting in reduced TGFβ transactivation. In contrast, TNF-α and IL-1β increased ανβ6 expression in both pKCs and epidermal KCs, resulting in increased TGFβ transactivation. Thus, loss of TGFβ transactivation is associated with homeostatic and UVB-induced LC migration but not with inflammation-induced migration.

      Results

      Enforced intrinsic TGFβ signaling is sufficient to prevent LC migration

      We previously reported that ligand-independent TGFβR signaling in LCs was sufficient to prevent LC migration during homeostasis and in response to UVB (
      • Mohammed J.
      • Beura L.K.
      • Bobr A.
      • Astry B.
      • Chicoine B.
      • Kashem S.W.
      • et al.
      Stromal cells control the epithelial residence of DCs and memory T cells by regulated activation of TGF-β.
      ). To determine whether LC migration in response to other inflammatory stimuli could also be overcome by ligand-independent TGFβR signaling, we utilized our hulangerinCreERT2 × lox-stop-lox TGFβRCA × ROSA26.lox-stop-lox-YFP mice (LCTGFβRCA). In these mice, tamoxifen treatment results in dual expression of YFP and a ligand-independent, constitutively active form TGFβR1 selectively in LCs. To test whether LC migration in response to inflammatory cytokines was affected by enforced TGFβR1 signaling, we treated LCTGFβRCA and control LCYFP mice with an intraperitoneal injection of tamoxifen for 5 days, followed by intradermal injection with 100 μg of TNF-α or PBS. The number of LCs in the epidermis at 72 hours after injection was determined by immunofluorescent microscopic evaluation of epidermal whole mounts. The number of LCs in the LNs was evaluated by flow cytometry gating LCs as CD11c+, major histocompatibility complex (MHC) IIhigh, CD11b+, CD207+, CD103. As expected, TNF-α induced an approximately 50% reduction in the number of epidermal LCs in control mice, but the numbers in LCTGFβRCA mice were unaffected (Figure 1a and c). Consistent with the known capacity for TNF-α to induce LC migration, we observed increased numbers of LCs in the LNs of TNF-α‒treated mice compared with that in the LNs of PBS-treated control mice (Figure 1d). Notably, LC numbers in the LNs of LCTGFβRCA mice were unaffected by TNF-α administration. We repeated the experiment with an intradermal injection of IL-1β (100 μg) and observed a similar result (Figure 1b, e, and f). Thus, injection of either TNF-α or IL-1β is sufficient to induce LC migration from the epidermis to the draining LNs, and this migration can be inhibited by conditionally expressing a constitutively active form of TGFβR in LCs.
      Figure thumbnail gr1
      Figure 1Enforce intrinsic TGFβ signaling is sufficient to prevent TNF-α‒ or IL-1β‒induced LC migration. LCTGFβRCA or control LCYFP mice received five-daily doses of tamoxifen (0.05 mg/g) to induce constitutive TGFβ signaling in LCs. The ear pinnae of these mice were subject to intradermal injection of either PBS, TNF-α, or IL-1β (100 ng). (a, b) Representative immunofluorescence and (c, e) summary data of epidermal sheets stained for MHCII (green) 72 hours after treatment with TNF-α or IL-1β; autofluorescence shows hair shafts. (d, f) The absolute number of LCs in dLNs after TNF-α or IL-1β treatment. The number of LCs in the LN was evaluated by flow cytometry gating LCs as CD11c+, MHC-IIhigh, CD11b+, CD207+, CD103. Data are representative of (d, f) two or at least (a‒c, e) three separate experiments. Each symbol represents data from an individual mouse. ∗P < 0.05 and ∗∗∗∗P < 0.0001. Bar = 100 μm. dLN, draining lymph node; LC, Langerhans cell; LN, lymph node; MHC, major histocompatibility complex.
      Epicutaneous infection with C. albicans drives the migration of LCs and dermal LCs from the skin into the draining LNs. To test whether constitutive TGFβR1 signaling can inhibit LC migration during C. albicans infection, we infected LCTGFβRCA and control LCYFP mice. Although the visualization of epidermal LCs in epidermal whole mounts was technically not feasible during C. albicans infection, we did observe increased numbers of LCs in the draining LNs of the control mice 3 days after infection. LC numbers were not increased in LCTGFβRCA mice, demonstrating an absence of efficient LC migration (Figure 2a). Notably, migration of cDC1 (CD11c+, MHC-IIhigh, CD207+, CD103+, CD11b), cDC2 (CD11c+, MHC-IIhigh, CD207, CD103, CD11b+), and double negative DC (CD11c+, MHC-IIhigh, CD207, CD103, CD11b) were equivalent in both control and LCTGFβRCA mice, thereby demonstrating that the failure to migrate in LCTGFβRCA mice is selective to LCs (Figure 2b‒d). Similar results also were obtained after a single epicutaneous application of 0.1 mM 7,12-dimethylbenz[a]anthracene (DMBA) (Figure 2e). In summary, these findings show that conditional expression of the constitutively active form of TGFβ receptor in LCs is sufficient to prevent both steady-state and inflammation-induced LC migration.
      Figure thumbnail gr2
      Figure 2Migration of dermal DCs is unaffected in TGFβRCALC mice. LCTGFβRCA or control LCYFP mice were treated with an intraperitoneal injection of tamoxifen as described earlier. Flank skin of mice was then treated with either epicutaneous Candida albicans or DMBA and PBS or vehicle controls, respectively. (a, e) Inguinal LCs were identified by flow cytometry by gating as CD11c+, MHC-IIhigh, CD11b+, CD207+, CD103 Different cDC subsets from C. albicans‒infected mice were detected by gating for cDC1 as CD11c+, MHC-IIhigh, CD207+, CD103+, and CD11b; cDN2 as CD11c+, MHC-IIhigh, CD207, CD103, and CD11b+; and dnDC as CD11c+, MHC-IIhigh, CD207, CD103, and CD11b. (a‒e) Data are representative of two separate experiments. Each symbol represents data from an individual mouse. ∗P < 0.05 and ∗∗∗∗P < 0.0001. cDC, cutaneous dendritic cell; DC, dendritic cell; DMBA, 7,12-dimethylbenz[a]anthracene; dnDC, double negative dendritic cell; LC, Langerhans cell; MHC, major histocompatibility complex; ns, not significant.

      Migratory stimuli increase integrin ανβ6 and active TGFβ bioavailability in pKCs

      We previously reported that exposure of KC to UVB reduced the expression of the TGFβ-activating integrins Itgb6 and Itgb8 in vivo and in vitro (
      • Mohammed J.
      • Beura L.K.
      • Bobr A.
      • Astry B.
      • Chicoine B.
      • Kashem S.W.
      • et al.
      Stromal cells control the epithelial residence of DCs and memory T cells by regulated activation of TGF-β.
      ). Coupled with the observation that forced TGFβR signaling prevented UVB-induced LC migration, we concluded that LC migration in response to UVB resulted from reduced epidermal bioavailability of TGFβ. To test whether inflammatory stimuli also suppress KC expression of Itgb6 and Itgb8, we treated in vitro pKC cultures with UVB, TNF-α, IL-1β, and DMBA and evaluated mRNA expression by RT-qPCR 24 hours after treatment. As expected, the expression of Itgb6 and Itgb8 were reduced by UVB exposure (Figure 3a and b). Similarly, expression of transcripts for the integrin-associated proteins Itgav (αν), Tln1 (Talin1), and Fermt1 (Kindlin1) was reduced, suggesting a broad reduction in the pathway responsible for transactivating latent TGFβ (Figure 3c‒e). Contrary to our expectations, we observed that pKC cultures incubated with TNF-α, IL-1β, or DMBA consistently increased the expression of Itgb6 (Figure 3a). Expression of Itgav, Tln1, and Fermt1 remained relatively unchanged, and expression of Itgb8 was reduced only with DMBA treatment (Figure 3b‒e). Analysis of protein expression by flow cytometry revealed reduced expression of ανβ6 after UVB treatment and modest but statistically significant increased expression after TNF-α and DMBA treatment (Figure 3f). Expression of ανβ8 was largely unchanged, with only a minor decrease after DMBA treatment (Figure 3g). Finally, we tested the capacity of pKCs to activate latent TGFβ by incubating UVB-, TNF-α‒, IL-1β‒, and DMBA-treated pKCs with a reporter cell line that expresses luciferase when autocrine TGFβ is transactivated by treated cells. Consistent with the expression data, pKCs treated with UVB showed reduced capacity to transactivate TGFβ, whereas pKCs treated with TNF-α, IL-1β , or DMBA showed enhanced TGFβ transactivation (Figure 3h). From these data, we conclude that the three inflammatory stimuli we have examined increase the expression of integrin ανβ6 and the capacity of in vitro pKCs to transactivate TGFβ.
      Figure thumbnail gr3
      Figure 3Migratory stimuli increase the ability of pKCs to convert latent TGFβ. pKCs were treated with 100 ngs of TNF-α and IL-1β or 0.1 mM DMBA for 24 hours or 20 mj UVB. (a‒e) RT-qPCR analysis of Itgb6, Itgb8, Itgav, Fermt1, and Tln1 is shown. (f, g) Flow cytometry analysis of the expression of integrins ανβ6 and ανβ8 on treated pKCs. Luciferase activity in tMLEC TGFβ reporter cells cocultured with pKCs pretreated with the indicated stimulus is shown. (h) Data are presented as RLU, normalized to control cells; PBS or vehicle-treated pKCs. Data are representative of (a‒e) two independent experiments or (h) three independent experiments. (a‒h) Each symbol represents data from an individual mouse. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, and ∗∗∗∗P < 0.0001. MFI, mean fluorescence intensity. DMBA, 7,12-dimethylbenz[a]anthracene; pKC, primary keratinocyte; RLU, relative light unit; tMLEC, thymic mink lung epithelial reporter cell.

      KC expression of Itgb6 is increased by TNF-α in vivo

      We have previously shown that subsets of KCs express different amounts of Itgb6 and Itgb8. In the steady state, IFE KCs primarily express Itgb6 and not Itgb8, IM KCs primarily express Itgb8 and not Itgb6, and bulge KCs express both (
      • Mohammed J.
      • Beura L.K.
      • Bobr A.
      • Astry B.
      • Chicoine B.
      • Kashem S.W.
      • et al.
      Stromal cells control the epithelial residence of DCs and memory T cells by regulated activation of TGF-β.
      ). We noted that pKCs expressed high levels of both Itgb6 and Itgb8, suggesting that pKCs may be more representative of bulge KCs rather than IFE or IM KCs. Because LCs are actively excluded from the bulge and reside in the IFE and IM (
      • Nagao K.
      • Kobayashi T.
      • Moro K.
      • Ohyama M.
      • Adachi T.
      • Kitashima D.Y.
      • et al.
      Stress-induced production of chemokines by hair follicles regulates the trafficking of dendritic cells in skin.
      ), we next examined the expression of Itgb6 and Itgb8 in IFE and IM KCs by RT-qPCR. Wild-type mice were injected intradermally with 100 μg of TNF-α or PBS on flank skin. After 4 or 24 hours, single-cell epidermal suspensions were FACS sorted as IFE (CD45.2, CD207, MHCII, CD34, epithelial cell adhesion molecule [EpCAM], Sca1+) or IM (CD45.2, CD207, MHCII, CD34, EpCAM+, Sca1). As expected, IFE KCs expressed higher levels of Itgb6 than IM KCs (Figure 4a). Administration of TNF-α transiently increased the expression of Itgb6 at 4 hours that returned to baseline level by 24 hours in IFE KCs. Somewhat unexpectedly, expression of Itgb6 also increased in IM KCs with similar kinetics. Expression of Itgb8 was limited to IM KCs in PBS-injected mice (Figure 4b). Administration of TNF-α resulted in a transient nonsignificant reduction of Itgb8 expression in IM KCs that returned to baseline level by 24 hours. Because the expression of Itgb8 only trended downward in TNF-α‒treated IM KCs, we examined whether TNF-α might preferentially drive the migration of LCs out of the follicular ostia. TNF-α was administered intradermally, and flank skin was harvested 48 hours later to visualize LCs by immunofluorescence. Follicular isthmus and LCs were identified by staining with α-EpCAM and α-Langerin, respectively. We observed equivalent LC migration from both the IFE and the IM (Figure 4c). Taken together, these data show that TNF-α transiently increased the expression of Itgb6 mRNA in IFE and IM KCs in vivo, whereas the expression of Itgb8 remained largely unchanged.
      Figure thumbnail gr4
      Figure 4TNF-α does not suppress integrin Itgb6 and Itgb8 mRNA expression by KC in vivo. RT-qPCR analysis of Itgb6 and Itgb8 mRNA from wild-type mouse sorted epidermal cells that had been treated with an intradermal injection. (a, b) TNF-α for 4 or 24 h. IFE KCs were gated as CD45.2, CD207, MHCII, CD34, EpCAM, and Sca1+ and IM KCs as CD45.2, CD207, MHCII, CD34, EpCAM+, and Sca1; (a, b) results are presented relative to Gapdh. (c) Microscopy of flank skin transverse sections from WT mice for 72 h after intradermal injection. TNF-α (100 ng) or PBS stained with Langerin (red), EpCAM (cyan), and DAPI. Data are representative of (a, b) two independent experiments or (c) three independent experiments. Bar = 50 μm. (a, b) Each symbol represents data from an individual mouse. ∗∗P < 0.01. EpCAM, epithelial cell adhesion molecule; h, hour; IFE, interfollicular; IM, infundibulum/isthmus; KC, keratinocyte; MHC, major histocompatibility complex; ns, not significant; Rel Expr, relative expression; WT, wild type.

      TNF-α increases epidermal TGFβ bioavailability through integrin ανβ6

      To confirm our finding that Itgb6 mRNA is increased in TNF-α treated IFE and IM KCs, we examined the surface expression of integrins ανβ6 and ανβ8 proteins by flow cytometry using recently developed mAbs (
      • Takasaka N.
      • Seed R.I.
      • Cormier A.
      • Bondesson A.J.
      • Lou J.
      • Elattma A.
      • et al.
      Integrin αvβ8-expressing tumor cells evade host immunity by regulating TGF-β activation in immune cells.
      ;
      • Weinreb P.H.
      • Simon K.J.
      • Rayhorn P.
      • Yang W.J.
      • Leone D.R.
      • Dolinski B.M.
      • et al.
      Function-blocking integrin alphavbeta6 monoclonal antibodies: distinct ligand-mimetic and nonligand-mimetic classes.
      ). In PBS-injected mice, we observed modest expression of ανβ6 by IFE KCs that was less evident in IM KCs (Figure 5a and b). Staining of KCs isolated from Itgb6–/– Itgb8ΔKC mice was included as a specificity control. Treatment with TNF-α increased the expression of ανβ6 in IFE KCs at 12 hours, which largely persisted at 24 hours. IM KCs also increased the expression of ανβ6, although less robustly than IFE KCs. Expression of ανβ8 was evident in PBS-treated IM KCs and modestly increased after TNF-α administration. We then tested whether the increased expression of ανβ6 on KCs resulted in increased transactivation of TGFβ. TNF-α or PBS was administered intradermally to wild-type mice, and bulk epidermal cells were isolated 12 hours later and incubated in vitro with a reporter cell line for active TGFβ, as described earlier. Epidermal cells from TNF-α‒treated skin showed an enhanced capacity to transactivate latent TGFβ compared with epidermal cells from PBS-treated skin (Figure 5c). Finally, to determine whether increased expression of integrin β6 is sufficient to increase the surface expression of ανβ6 and increase TGFβ activation, we overexpressed integrin β6 in a cell line generated from immortalized primary murine KCs. Cells were electroporated with either an empty vector or a vector containing Itgb6 under the control of the cytomegalovirus promoter (β6), followed by antibiotic selection and cloning by limited dilution. The surface expression of integrin ανβ6 was clearly increased in KCs transfected with Itgb6 compared with that transfected with the empty vector (Figure 5d). As expected, KCs transfected with Itgb6 showed increased activation of latent TGFβ (Figure 5d). Thus, increased integrin β6 expression is sufficient to increase TGFβ activation. These data show that overexpression of integrin β6 in KCs or administration of TNF-α in the skin drives increased expression of ανβ6 and ανβ8 and that this is sufficient to increase the epidermal bioavailability of active TGFβ.
      Figure thumbnail gr5
      Figure 5TNF-α increases the surface expression of integrins αvβ6 and ανβ8 on epidermal KCs. Representative flow plots of integrins ανβ6 and ανβ8 on (a) IFE and (b) IM KC populations from WT or Itgb6–/– Itgb8ΔKC mice that had received the intradermal injection. TNF-α (100 ng) for 12 or 24 h; (a) IFE KCs were gated as CD45.2, CD207, MHCII, CD34, EpCAM, and Sca1+, and (b) IM KCs were gated as CD45.2, CD207, MHCII, CD34, EpCAM+, and Sca1. Luciferase activity in tMLEC TGFβ reporter cells cocultured with bulk epidermal cells from WT mice was harvested 24 h after in vivo intradermal injection of TNF-α (100 ng); (c) data are presented as RLU. Integrin ανβ6 expression in a KC cell line transfected with an EV or with an integrin β6 plasmid construct (β6). (d) Control-indicated EV cells stained with FMO. Luciferase activity in tMLEC TGFβ reporter cells cocultured with either EV or β6 transfected KC cell line. (e) Flow cytometry analysis of ανβ6 expression on KC cells that been transfected with either EV or β6 vector. Data are presented as RLU, normalized to control cells. (a, b, e) Data are representative of two independent experiments with a cohort size of 3‒4. (c) Each symbol represents data from an individual mouse. ∗P < 0.05. EpCAM, epithelial cell adhesion molecule; EV, empty plasmid construct; h, hour; FMO, fluorescence minus 1; IFE, interfollicular; IM, infundibulum/isthmus; KC, keratinocyte; MHC, major histocompatibility complex; RLU, relative light unit; tMLEC, thymic mink lung epithelial reporter cell; WT, wild type.

      Discussion

      In this study, we have shown that enforced TGFβR signaling in LCs is sufficient to prevent inflammation-induced LC migration. This extends our previous findings that homeostatic and UVB-induced LC migration can be inhibited by TGFβRCA expression. We also confirmed our previous observation that UVB reduced the expression of ανβ6 and ανβ8 in pKCs, resulting in reduced TGFβ transactivation. However, contrary to our expectations, TNF-α and IL-1β did not decrease but rather increased Itgb6 and ανβ6 expression in both pKCs and epidermal KCs. This was associated with an increased KC-mediated TGFβ transactivation. Thus, loss of TGFβ transactivation by KCs is associated with homeostatic and UVB-induced LC migration but not with inflammation-induced migration.
      We had expected that loss of TGFβ signaling through the suppression of KC-mediated transactivation would be a fundamental component in LC migration. The observation that forced expression of TGFβRCA can prevent LC migration indicates that high levels of TGFβR signaling can overcome a wide variety of migratory signals. The fact that we did not observe any reduction in KC-mediated TGFβ transactivation in response to TNF-α or IL-1β indicates that these signals do not induce LC migration through an extrinsic loss of TGFβ availability. TNF-α‒induced LC migration is inhibited by global ablation of TNF receptor 2 (p50) but not of TNF receptor 1 (p75) (
      • Wang B.
      • Fujisawa H.
      • Zhuang L.
      • Kondo S.
      • Shivji G.M.
      • Kim C.S.
      • et al.
      Depressed Langerhans cell migration and reduced contact hypersensitivity response in mice lacking TNF receptor p75.
      ;
      • Wang B.
      • Kondo S.
      • Shivji G.M.
      • Fujisawa H.
      • Mak T.W.
      • Sauder D.N.
      Tumour necrosis factor receptor II (p75) signalling is required for the migration of Langerhans’ cells.
      ). In the epidermis, TNF receptor 2 is believed to be preferentially expressed by LCs (
      • Eaton L.H.
      • Roberts R.A.
      • Kimber I.
      • Dearman R.J.
      • Metryka A.
      Skin sensitization induced Langerhans' cell mobilization: variable requirements for tumour necrosis factor-α.
      ;
      • Luo D.
      • Luo Y.
      • He Y.
      • Zhang H.
      • Zhang R.
      • Li X.
      • et al.
      Differential functions of tumor necrosis factor receptor 1 and 2 signaling in ischemia-mediated arteriogenesis and angiogenesis [published correction appears in Am J Pathol 2007;171:1078].
      ;
      • Yang S.
      • Wang J.
      • Brand D.D.
      • Zheng S.G.
      Role of TNF-TNF receptor 2 signal in regulatory T cells and its therapeutic implications.
      ). Thus, TNF-α could likely induce migration by acting directly on LCs. In contrast, LC migration in response to IL-1β is not affected by the loss of Myd88, which is a required component of IL-1β receptor signaling (
      • Didovic S.
      • Opitz F.V.
      • Holzmann B.
      • Förster I.
      • Weighardt H.
      Requirement of MyD88 signaling in keratinocytes for Langerhans cell migration and initiation of atopic dermatitis-like symptoms in mice.
      ;
      • Haley K.
      • Igyártó B.Z.
      • Ortner D.
      • Bobr A.
      • Kashem S.
      • Schenten D.
      • et al.
      Langerhans cells require MyD88-dependent signals for Candida albicans response but not for contact hypersensitivity or migration.
      ). Thus, IL-1β likely triggers LC migration through an LC-extrinsic mechanism that is independent of reduced TGFβ transactivation. The functional importance of increased ανβ6 expression in response to TNF-α and IL-1β remains unclear. TGFβ is required for differentiation of resident memory T cells and could participate in the development of monocyte-derived LCs that are recruited into the skin by inflammation (
      • Ferrer I.R.
      • West H.C.
      • Henderson S.
      • Ushakov D.S.
      • Santos E Sousa P.
      • Strid J.
      • et al.
      A wave of monocytes is recruited to replenish the long-term Langerhans cell network after immune injury.
      ;
      • Mackay L.K.
      • Rahimpour A.
      • Ma J.Z.
      • Collins N.
      • Stock A.T.
      • Hafon M.L.
      • et al.
      The developmental pathway for CD103(+)CD8+ tissue-resident memory T cells of skin.
      ). We speculate that increased ανβ6 expression may be required for optimal recruitment and differentiation of these cell types in an inflammatory context.
      It is interesting to note that the two conditions where the loss of KC-mediated TGFβ transactivation triggers LC migration (i.e., homeostasis and UVB) are both associated with the induction of peripheral tolerance (
      • Mutyambizi K.
      • Berger C.L.
      • Edelson R.L.
      The balance between immunity and tolerance: the role of Langerhans cells.
      ;
      • Shklovskaya E.
      • O'Sullivan B.J.
      • Ng L.G.
      • Roediger B.
      • Thomas R.
      • Weninger W.
      • et al.
      Langerhans cells are precommitted to immune tolerance induction.
      ;
      • Yoshiki R.
      • Kabashima K.
      • Sakabe J.
      • Sugita K.
      • Bito T.
      • Nakamura M.
      • et al.
      The mandatory role of IL-10-producing and OX40 ligand-expressing mature Langerhans cells in local UVB-induced immunosuppression.
      ). This is consistent with the observation that those LCs that have migrated into regional LNs in response to an artificial loss of autocrine TGFβ maintain an immature activation state (
      • Bobr A.
      • Igyarto B.Z.
      • Haley K.M.
      • Li M.O.
      • Flavell R.A.
      • Kaplan D.H.
      Autocrine/paracrine TGF-β1 inhibits Langerhans cell migration.
      ). Explorations of the genomic state of LCs after different types of migratory stimuli represent an exciting future avenue to explore this question.

      Materials and Methods

      Mice

      HuLangerin-CreERT2 and TGFβRCA mice have been previously described (
      • Bartholin L.
      • Cyprian F.S.
      • Vincent D.
      • Garcia C.N.
      • Martel S.
      • Horvat B.
      • et al.
      Generation of mice with conditionally activated transforming growth factor beta signaling through the TbetaRI/ALK5 receptor.
      ;
      • Bobr A.
      • Igyarto B.Z.
      • Haley K.M.
      • Li M.O.
      • Flavell R.A.
      • Kaplan D.H.
      Autocrine/paracrine TGF-β1 inhibits Langerhans cell migration.
      ). Itgb6−/− and Itgb8loxP mice were kindly provided by D. Sheppard (University of California, San Francisco). HuLangerin-CreERT2 mice were bred with TGFβRCA and ROSA26.LSL.YFP (Jackson Laboratory, Bar Harbor, ME) reporter mice, resulting in TGFβRCALC mice (
      • Mohammed J.
      • Beura L.K.
      • Bobr A.
      • Astry B.
      • Chicoine B.
      • Kashem S.W.
      • et al.
      Stromal cells control the epithelial residence of DCs and memory T cells by regulated activation of TGF-β.
      ). C57BL/6 (wild type) and Tg(KRT14-cre)1Amc/J (K14-Cre) mice were purchased from Jackson Laboratory. We crossed K14-Cre mice with Itgb8loxP and Itgb6−/− mice to obtain Itgb6−/− and Itgb8ΔKC mice (
      • Mohammed J.
      • Beura L.K.
      • Bobr A.
      • Astry B.
      • Chicoine B.
      • Kashem S.W.
      • et al.
      Stromal cells control the epithelial residence of DCs and memory T cells by regulated activation of TGF-β.
      ). We used age- and sex-matched mice that were aged between 6 and 12 weeks in all experiments. All mice were maintained under specific pathogen‒free conditions, and all mouse experiments were approved by the University of Pittsburgh Institutional Animal Care and Use Committee (Pittsburgh, PA).

      Reagents and treatments

      TNF-α (315-01A) or IL-1β (211-11B) recombinant mouse peptides were purchased from PeproTech (Rocky Hills, NJ) and were resuspended in PBS. For UVB experiments, we used two TL 20W/12RS lamps (Philips, Amsterdam, The Netherlands). We used UVB doses previously described as 20 mJ/cm2 for pKCs or experiments, (
      • Mohammed J.
      • Beura L.K.
      • Bobr A.
      • Astry B.
      • Chicoine B.
      • Kashem S.W.
      • et al.
      Stromal cells control the epithelial residence of DCs and memory T cells by regulated activation of TGF-β.
      ).
      DMBA (57-97-6; Sigma-Aldrich, St. Louis, MI) was applied at 0.1mM in DMSO for pKC experiments. For in vivo mouse experiments, 10 mM of DMBA was resuspended in DMSO-to-ethanol-to-glycerol solution (1:1:3) and was applied to the flank skin of shaven mice. Antibodies directly conjugated to different fluorophores were used for flow cytometry and immunofluorescence. Anti-CD11c(N418)-PerCp5.5, CD11b(M1/70)-PeCy7, I-A/I-E/MHCII (M5/114.15.2)-AF700, Langerin(4c7)-phycoerythrin, CD103(2E7)-AF647, CD45.2(104)-BV605, Sca-1(E13-161.7)-PerCp5.5, EpCAM(G8.8)-PeCy7, and CD34(HM34)-PE/Dazzle594 were purchased from BioLegend (San Diego, CA). Viability Dye eFluor 780 (eBioscience; Invitrogen, Carlsbad, CA) was used for live‒dead discrimination. Anti-ανβ6 (6.3g9) and anti-ανβ8 (C6D4)-phycoerythrin were kindly provided by Dean Sheppard and Stephen Nishimura, respectively. Anti-ανβ6 (6.3g9) was directly conjugated to Alexa Fluor 647 (A20186; Thermo Fischer Scientific, Waltham, MA).

      Tamoxifen treatment

      Tamoxifen (T5648; Sigma-Aldrich) was dissolved in 1/10 volume of 200 proof ethanol with repeated incubations at 55 °C (15–30 seconds) and vortexing. The tamoxifen‒ethanol mixture was diluted with corn oil (C8267; Sigma-Aldrich) to a final concentration of 10 mg/ml. Before treatments, TGFβRCALC mice received 5 consecutive days of intraperitoneal injection of tamoxifen at 0.05 mg/g of mouse weight.

      Immunofluorescence and imaging

      Epidermal sheets were prepared as previously described (
      • Mohammed J.
      • Beura L.K.
      • Bobr A.
      • Astry B.
      • Chicoine B.
      • Kashem S.W.
      • et al.
      Stromal cells control the epithelial residence of DCs and memory T cells by regulated activation of TGF-β.
      ). Briefly, skin fat was mechanically removed and subsequently mounted on microscopy slides that had been precoated with double-sided adhesive tape (3M, St. Paul, MN). Slides were incubated in 10 mM EDTA at 37 °C for 45–90 minutes. The dermis was peeled away from the epidermis with standard forceps. Skin-whole mounts were prepared from skin samples embedded in optimal cutting temperature compound, and 8 μm transverse skin slice sections were prepared. Epidermal sheets and skin-whole mounts were fixed in 4% paraformaldehyde at room temperature for 30 minutes and blocked for 1 hour at room temperature in PBS buffer containing 0.1% tween-20, 2% BSA, and 2% rat serum. Immunostaining of skin samples was done overnight in PBS containing 0.1% tween-20 and 0.5% BSA. Skin samples were stained with anti-EpCAM(G8.8)-AF647, anti-A/I-E/MHCII-AF488, and anti-Langerin(CD207)-phycoerythrin, followed by anti-phycoerythrin-AF555 and DAPI. Images were captured on an IX83 fluorescent microscope (Olympus, Tokyo, Japan) using a ×10 objective; image analysis was performed using cellSens Dimension software (Olympus).

      Flow cytometry

      Single-cell suspension from tissues was prepared as previously described (
      • Mohammed J.
      • Beura L.K.
      • Bobr A.
      • Astry B.
      • Chicoine B.
      • Kashem S.W.
      • et al.
      Stromal cells control the epithelial residence of DCs and memory T cells by regulated activation of TGF-β.
      ). Epidermal single-cell suspensions were prepared from shaved mouse skin that was incubated for 2 hours at 37 °C in 0.3% in 150 mM sodium chloride, 0.5 mM potassium chloride, and 0.5 mM glucose. The skin was minced finely with scissors and resuspended in RPMI 1640 media (Gibco, Grand Island, NY) containing 2.5 mg/ml Collagenase XI (Sigma-Aldrich), 0.1 mg/ml DNase (Sigma-Aldrich), 0.01 M 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (Sigma-Aldrich), and 10% fetal bovine serum, followed by incubation in a shaking incubator for 30 minutes at 37 °C. The resulting cell mesh was filtered through a 40-μm cell strainer (BD Biosciences, San Jose, CA). LNs (axillary and inguinal) were incubated in 400 U/ml Collagenase D (Roche Applied Science, Penzberg, Germany) and 0.1 mg/ml DNase in RPMI 1640 with 10% fetal bovine serum for 40 minutes at 37 °C and then minced through a 40-μm cell strainer. Single-cell suspensions were blocked with 2.4G2 culture supernatant (ATCC, Manassas, VA). Surface staining was performed in standard FACS buffer for 30 minutes at 4 °C. For intracellular cytokine staining of Langerin (CD207), cells were first fixed and permeabilized with Cytofix/Cytoperm kit (BD Biosciences) according to the manufacturer’s instructions. LSRFortessa flow cytometers (Becton Dickinson, Franklin Lakes, NJ) and Flowjo software (TreeStar, Ashland, OR) were used for analysis. Epidermal KC subsets were sorted on FACSAria cell sorter (BD Biosciences) and gated as IFE KCs: CD45.2, Sca1+, CD34, and EpCAM and IM KCs: CD45.2, Sca1, CD34, and EpCAM+.

      RT-qPCR

      Total RNA from flow cytometry–sorted epidermal cells and pKC cultures were extracted with Thermo Fischer's Trizol-LS (10296028, Carlsbad, CA) or RNeasy Mini extraction kit (Qiagen, Hilden, Germany) following the manufacturer’s instructions and was quantified using Nanodrop (NanoDrop Technologies, Wilmington, DE). cDNA was generated using a High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA) and was subjected to RT-qPCR using TaqMan Gene Expression Master Mix and TaqMan Gene Expression Assays for Gapdh, Itgb6, Itgb8, Itgav, Fermt1 (Kindlin1), and Tln1.

      Cell culture

      Newborn mice (aged 1–4 days) were used to culture pKCs as previously described (
      • Dlugosz A.A.
      • Glick A.B.
      • Tennenbaum T.
      • Weinberg W.C.
      • Yuspa S.H.
      Isolation and utilization of epidermal keratinocytes for oncogene research.
      ;
      • Mohammed J.
      • Beura L.K.
      • Bobr A.
      • Astry B.
      • Chicoine B.
      • Kashem S.W.
      • et al.
      Stromal cells control the epithelial residence of DCs and memory T cells by regulated activation of TGF-β.
      ). TMLECs that had been previously transfected with a plasmid containing the luciferase cDNA downstream of a TGFβ-responsive portion of the PAI-1 promoter were cultured as originally described (
      • Abe M.
      • Harpel J.G.
      • Metz C.N.
      • Nunes I.
      • Loskutoff D.J.
      • Rifkin D.B.
      An assay for transforming growth factor-beta using cells transfected with a plasminogen activator inhibitor-1 promoter-luciferase construct.
      ).

      KC transfection

      We have generated a stable KC cell line from pKCs by spontaneous immortalization. These cells were grown at 37 °C, at 5% carbon dioxide, and in EMEM media complemented with 8% chelex serum and 1% penicillin/streptomycin for 8 weeks, followed by limited dilution cloning. The resulting cell line was transfected with a pCDNA3.1+/C-(K)-DYK-ITGB6 (β6) plasmid construct or with an empty pCDNA3.1+/C-(K)-DYK (empty vector) that were purchased from GenScript Biotech (Piscataway, NJ). Transfections were performed in 4-mm-gap cuvettes using a BTX ECM 830 square wave electroporator. Transfection was accomplished by a single pulse of 300 volts and 10 ms, with 5‒10 μg of plasmid DNA per 1.0 × 106 cells to 2.0 × 106 cells. Four days after transfection, selection reagent G418 (0.1‒2.5 mg/ml; Sigma-Aldrich) was added to the medium to select the stably transfected cells for 15 days, with the medium being refreshed every other day. The cells were cloned by limiting dilution.

      TGFβ-activation reporter assay

      The ability of KCs to transactivate TGFβ was determined by coculture of KCs with thymic mink lung epithelial reporter cells as previously described (
      • Mohammed J.
      • Beura L.K.
      • Bobr A.
      • Astry B.
      • Chicoine B.
      • Kashem S.W.
      • et al.
      Stromal cells control the epithelial residence of DCs and memory T cells by regulated activation of TGF-β.
      ). Briefly, thymic mink lung epithelial reporter cells were grown in DMEM media containing 10% fetal calf serum and were plated at 2.0 × 104 cells per well in a 96-well cell culture‒treated plate for 3 hours at 37 °C and 5% carbon dioxide. TNF-α, IL-1β, UVB, DMBA, or nontreated KCs were harvested with 0.25% trypsin and 2.21 mM EDTA (25 053-Cl; Corning, Corning, NY) for 5 minutes at 37 °C and 5% carbon dioxide. Experimental pKCs were cocultured with thymic mink lung epithelial reporter cells at a density of 4.0 × 104 cells in 1% chelexed serum EMEM media as previously described (
      • Abe M.
      • Harpel J.G.
      • Metz C.N.
      • Nunes I.
      • Loskutoff D.J.
      • Rifkin D.B.
      An assay for transforming growth factor-beta using cells transfected with a plasminogen activator inhibitor-1 promoter-luciferase construct.
      ;
      • Dlugosz A.A.
      • Glick A.B.
      • Tennenbaum T.
      • Weinberg W.C.
      • Yuspa S.H.
      Isolation and utilization of epidermal keratinocytes for oncogene research.
      ). The cells were cultured for 16–20 hours, after which the reporter cells were lysed and assayed for luciferase activity using Bright Glo Luciferase Assay System (Promega, Madison, WI).

      Statistical analysis

      Groups were compared with Prism software (GraphPad Software, San Diego, CA) using the two-tailed unpaired Student’s t-test. Data are presented as mean only or mean ± SEM. P < 0.05 was considered significant.

      Data availability statement

      No datasets were generated or analyzed during this study.

      ORCIDs

      Jacinto S. De La Cruz Diaz: http://orcid.org/0000-0003-4453-0014
      Breanna Anh-Thu Nguyen: http://orcid.org/0000-0001-8642-6026

      Author Contributions

      Conceptualization: JSDLCD, TH, DHK; Formal Analysis: JSDLCD; Funding Acquisition: DHK; Investigation: JSDLCD, YY, YZ, BATN, HL; Methodology: TH, JSDLCD; Project Administration: DHK; Resources: SN; Supervision: DHK; Validation: JSDLCD; Visualization: JSDLCD, DHK; Writing – Original Draft Preparation: JSDLCD, DHK; Writing – Review and Editing: JSDLCD, DHK

      Conflict of Interest

      The authors state no conflict of interest.

      Acknowledgments

      We thank the members of the Kaplan Laboratory and members of the Universiy of Pittsburgh, departments of Dermatology and Immunology for helpful discussions. This work was supported by National Institutes of Health grants 5T32AI089443 (JSDLCD) and 5R01AR060744 (DHK and JSDLCD).

      References

        • Abe M.
        • Harpel J.G.
        • Metz C.N.
        • Nunes I.
        • Loskutoff D.J.
        • Rifkin D.B.
        An assay for transforming growth factor-beta using cells transfected with a plasminogen activator inhibitor-1 promoter-luciferase construct.
        Anal Biochem. 1994; 216: 276-284
        • Aluwihare P.
        • Mu Z.
        • Zhao Z.
        • Yu D.
        • Weinreb P.H.
        • Horan G.S.
        • et al.
        Mice that lack activity of alphavbeta6- and alphavbeta8-integrins reproduce the abnormalities of Tgfb1- and Tgfb3-null mice.
        J Cell Sci. 2009; 122: 227-232
        • Bartholin L.
        • Cyprian F.S.
        • Vincent D.
        • Garcia C.N.
        • Martel S.
        • Horvat B.
        • et al.
        Generation of mice with conditionally activated transforming growth factor beta signaling through the TbetaRI/ALK5 receptor.
        Genesis. 2008; 46: 724-731
        • Bobr A.
        • Igyarto B.Z.
        • Haley K.M.
        • Li M.O.
        • Flavell R.A.
        • Kaplan D.H.
        Autocrine/paracrine TGF-β1 inhibits Langerhans cell migration.
        Proc Natl Acad Sci USA. 2012; 109: 10492-10497
        • Borkowski T.A.
        • Letterio J.J.
        • Farr A.G.
        • Udey M.C.
        A role for endogenous transforming growth factor beta 1 in Langerhans cell biology: the skin of transforming growth factor beta 1 null mice is devoid of epidermal Langerhans cells.
        J Exp Med. 1996; 184: 2417-2422
        • Cumberbatch M.
        • Dearman R.J.
        • Kimber I.
        Langerhans cells require signals from both tumour necrosis factor-alpha and interleukin-1 beta for migration.
        Immunology. 1997; 92: 388-395
        • Cumberbatch M.
        • Fielding I.
        • Kimber I.
        Modulation of epidermal Langerhans' cell frequency by tumour necrosis factor-alpha.
        Immunology. 1994; 81: 395-401
        • Didovic S.
        • Opitz F.V.
        • Holzmann B.
        • Förster I.
        • Weighardt H.
        Requirement of MyD88 signaling in keratinocytes for Langerhans cell migration and initiation of atopic dermatitis-like symptoms in mice.
        Eur J Immunol. 2016; 46: 981-992
        • Dlugosz A.A.
        • Glick A.B.
        • Tennenbaum T.
        • Weinberg W.C.
        • Yuspa S.H.
        Isolation and utilization of epidermal keratinocytes for oncogene research.
        Methods Enzymol. 1995; 254: 3-20
        • Eaton L.H.
        • Roberts R.A.
        • Kimber I.
        • Dearman R.J.
        • Metryka A.
        Skin sensitization induced Langerhans' cell mobilization: variable requirements for tumour necrosis factor-α.
        Immunology. 2015; 144: 139-148
        • Ferrer I.R.
        • West H.C.
        • Henderson S.
        • Ushakov D.S.
        • Santos E Sousa P.
        • Strid J.
        • et al.
        A wave of monocytes is recruited to replenish the long-term Langerhans cell network after immune injury.
        Sci Immunol. 2019; 4eaax8704
        • Haley K.
        • Igyártó B.Z.
        • Ortner D.
        • Bobr A.
        • Kashem S.
        • Schenten D.
        • et al.
        Langerhans cells require MyD88-dependent signals for Candida albicans response but not for contact hypersensitivity or migration.
        J Immunol. 2012; 188: 4334-4339
        • Hemmi H.
        • Yoshino M.
        • Yamazaki H.
        • Naito M.
        • Iyoda T.
        • Omatsu Y.
        • et al.
        Skin antigens in the steady state are trafficked to regional lymph nodes by transforming growth factor-beta1-dependent cells.
        Int Immunol. 2001; 13: 695-704
        • Igyártó B.Z.
        • Haley K.
        • Ortner D.
        • Bobr A.
        • Gerami-Nejad M.
        • Edelson B.T.
        • et al.
        Skin-resident murine dendritic cell subsets promote distinct and opposing antigen-specific T helper cell responses.
        Immunity. 2011; 35: 260-272
        • Kaplan D.H.
        Ontogeny and function of murine epidermal Langerhans cells.
        Nat Immunol. 2017; 18: 1068-1075
        • Kaplan D.H.
        • Li M.O.
        • Jenison M.C.
        • Shlomchik W.D.
        • Flavell R.A.
        • Shlomchik M.J.
        Autocrine/paracrine TGFbeta1 is required for the development of epidermal Langerhans cells.
        J Exp Med. 2007; 204: 2545-2552
        • Kel J.M.
        • Girard-Madoux M.J.
        • Reizis B.
        • Clausen B.E.
        TGF-beta is required to maintain the pool of immature Langerhans cells in the epidermis.
        J Immunol. 2010; 185: 3248-3255
        • Kerr B.A.
        • Byzova T.V.
        Integrin alpha V (ITGAV).
        in: Choi S. Encyclopedia of signaling molecules. Second edition. Springer International Publishing, Cham, Switzerland2018: 2634-2645
        • Kobayashi T.
        • Glatz M.
        • Horiuchi K.
        • Kawasaki H.
        • Akiyama H.
        • Kaplan D.H.
        • et al.
        Dysbiosis and Staphylococcus aureus colonization drives inflammation in atopic dermatitis.
        Immunity. 2015; 42: 756-766
        • Kobayashi T.
        • Naik S.
        • Nagao K.
        Choreographing immunity in the skin epithelial barrier.
        Immunity. 2019; 50: 552-565
        • Luo D.
        • Luo Y.
        • He Y.
        • Zhang H.
        • Zhang R.
        • Li X.
        • et al.
        Differential functions of tumor necrosis factor receptor 1 and 2 signaling in ischemia-mediated arteriogenesis and angiogenesis [published correction appears in Am J Pathol 2007;171:1078].
        Am J Pathol. 2006; 169: 1886-1898
        • Mackay L.K.
        • Rahimpour A.
        • Ma J.Z.
        • Collins N.
        • Stock A.T.
        • Hafon M.L.
        • et al.
        The developmental pathway for CD103(+)CD8+ tissue-resident memory T cells of skin.
        Nat Immunol. 2013; 14: 1294-1301
        • Mohammed J.
        • Beura L.K.
        • Bobr A.
        • Astry B.
        • Chicoine B.
        • Kashem S.W.
        • et al.
        Stromal cells control the epithelial residence of DCs and memory T cells by regulated activation of TGF-β.
        Nat Immunol. 2016; 17: 414-421
        • Mutyambizi K.
        • Berger C.L.
        • Edelson R.L.
        The balance between immunity and tolerance: the role of Langerhans cells.
        Cell Mol Life Sci. 2009; 66: 831-840
        • Nagao K.
        • Kobayashi T.
        • Moro K.
        • Ohyama M.
        • Adachi T.
        • Kitashima D.Y.
        • et al.
        Stress-induced production of chemokines by hair follicles regulates the trafficking of dendritic cells in skin.
        Nat Immunol. 2012; 13: 744-752
        • Qin J.
        • Vinogradova O.
        • Plow E.F.
        Integrin bidirectional signaling: a molecular view.
        PLoS Biol. 2004; 2: e169
        • Shklovskaya E.
        • O'Sullivan B.J.
        • Ng L.G.
        • Roediger B.
        • Thomas R.
        • Weninger W.
        • et al.
        Langerhans cells are precommitted to immune tolerance induction.
        Proc Natl Acad Sci USA. 2011; 108: 18049-18054
        • Takada Y.
        • Ye X.
        • Simon S.
        The integrins.
        Genome Biol. 2007; 8: 215
        • Takasaka N.
        • Seed R.I.
        • Cormier A.
        • Bondesson A.J.
        • Lou J.
        • Elattma A.
        • et al.
        Integrin αvβ8-expressing tumor cells evade host immunity by regulating TGF-β activation in immune cells.
        JCI Insight. 2018; 3e122591
        • Travis M.A.
        • Sheppard D.
        TGF-β activation and function in immunity.
        Annu Rev Immunol. 2014; 32: 51-82
        • Wang B.
        • Fujisawa H.
        • Zhuang L.
        • Kondo S.
        • Shivji G.M.
        • Kim C.S.
        • et al.
        Depressed Langerhans cell migration and reduced contact hypersensitivity response in mice lacking TNF receptor p75.
        J Immunol. 1997; 159: 6148-6155
        • Wang B.
        • Kondo S.
        • Shivji G.M.
        • Fujisawa H.
        • Mak T.W.
        • Sauder D.N.
        Tumour necrosis factor receptor II (p75) signalling is required for the migration of Langerhans’ cells.
        Immunology. 1996; 88: 284-288
        • Weinreb P.H.
        • Simon K.J.
        • Rayhorn P.
        • Yang W.J.
        • Leone D.R.
        • Dolinski B.M.
        • et al.
        Function-blocking integrin alphavbeta6 monoclonal antibodies: distinct ligand-mimetic and nonligand-mimetic classes.
        J Biol Chem. 2004; 279: 17875-17887
        • Worthington J.J.
        • Klementowicz J.E.
        • Travis M.A.
        TGFβ: a sleeping giant awoken by integrins.
        Trends Biochem Sci. 2011; 36: 47-54
        • Yang S.
        • Wang J.
        • Brand D.D.
        • Zheng S.G.
        Role of TNF-TNF receptor 2 signal in regulatory T cells and its therapeutic implications.
        Front Immunol. 2018; 9: 784
        • Yang Z.
        • Mu Z.
        • Dabovic B.
        • Jurukovski V.
        • Yu D.
        • Sung J.
        • et al.
        Absence of integrin-mediated TGFbeta1 activation in vivo recapitulates the phenotype of TGFbeta1-null mice.
        J Cell Biol. 2007; 176: 787-793
        • Yoshiki R.
        • Kabashima K.
        • Sakabe J.
        • Sugita K.
        • Bito T.
        • Nakamura M.
        • et al.
        The mandatory role of IL-10-producing and OX40 ligand-expressing mature Langerhans cells in local UVB-induced immunosuppression.
        J Immunol. 2010; 184: 5670-5677