Advertisement

RXRα Ablation in Epidermal Keratinocytes Enhances UVR-Induced DNA Damage, Apoptosis, and Proliferation of Keratinocytes and Melanocytes

  • Zhixing Wang
    Affiliations
    Department of Pharmaceutical Sciences, College of Pharmacy, Oregon State University, Corvallis, Oregon, USA
    Search for articles by this author
  • Daniel J. Coleman
    Affiliations
    Department of Pharmaceutical Sciences, College of Pharmacy, Oregon State University, Corvallis, Oregon, USA

    Molecular and Cellular Biology Program, Oregon State University, Corvallis, Oregon, USA
    Search for articles by this author
  • Gaurav Bajaj
    Affiliations
    Department of Pharmaceutical Sciences, College of Pharmacy, Oregon State University, Corvallis, Oregon, USA
    Search for articles by this author
  • Xiaobo Liang
    Affiliations
    Department of Pharmaceutical Sciences, College of Pharmacy, Oregon State University, Corvallis, Oregon, USA
    Search for articles by this author
  • Gitali Ganguli-Indra
    Affiliations
    Department of Pharmaceutical Sciences, College of Pharmacy, Oregon State University, Corvallis, Oregon, USA
    Search for articles by this author
  • Arup K. Indra
    Correspondence
    Department of Pharmaceutical Sciences, College of Pharmacy, Oregon State University, 203 Pharmacy Building, 1601 SW Jefferson Avenue, Corvallis, Oregon 97331, USA
    Affiliations
    Department of Pharmaceutical Sciences, College of Pharmacy, Oregon State University, Corvallis, Oregon, USA

    Molecular and Cellular Biology Program, Oregon State University, Corvallis, Oregon, USA

    Environmental Health Sciences Center, Oregon State University, Corvallis, Oregon, USA

    Department of Dermatology, Oregon Health and Science University, Portland, Oregon, USA
    Search for articles by this author
      We show here that keratinocytic nuclear receptor retinoid X receptor-α (RXRα) regulates mouse keratinocyte and melanocyte homeostasis following acute UVR. Keratinocytic RXRα has a protective role in UVR-induced keratinocyte and melanocyte proliferation/differentiation, oxidative stress-mediated DNA damage, and cellular apoptosis. We discovered that keratinocytic RXRα, in a cell-autonomous manner, regulates mitogenic growth responses in skin epidermis through secretion of heparin-binding EGF-like growth factor, GM-CSF, IL-1α, and cyclooxygenase-2 and activation of mitogen-activated protein kinase pathways. We identified altered expression of several keratinocyte-derived mitogenic paracrine growth factors such as endothelin 1, hepatocyte growth factor, α-melanocyte stimulating hormone, stem cell factor, and fibroblast growth factor-2 in skin of mice lacking RXRα in epidermal keratinocytes (RXRαep-/- mice), which in a non-cell-autonomous manner modulated melanocyte proliferation and activation after UVR. RXRαep-/- mice represent a unique animal model in which UVR induces melanocyte proliferation/activation in both epidermis and dermis. Considered together, the results of our study suggest that RXR antagonists, together with inhibitors of cell proliferation, can be effective in preventing solar UVR-induced photocarcinogenesis.

      Abbreviations

      COX-2
      cyclooxygenase-2
      CPD
      cyclobutane pyrimidine dimers
      DAPI
      4′,6-diamidino-2-phenyl indole
      ET-1
      endothelin 1
      IHC
      immunohistochemistry
      8-oxo-dG
      8-hydroxy-2′-deoxyguanosine
      RXRα
      retinoid X receptor-α
      SCF
      stem cell factor

      Introduction

      Skin—the largest organ of the body—protects the organism from environmental, physical, and chemical traumas (
      • Lippens S.
      • Hoste E.
      • Vandenabeele P.
      • et al.
      Cell death in the skin.
      ). It is composed of the epidermis and mesenchyme-derived dermis, which are separated by a basement membrane. The epidermis is made up mainly of keratinocytes and composed of four layers: basal, spinous, granular, and stratum corneum (
      • Fuchs E.
      • Raghavan S.
      Getting under the skin of epidermal morphogenesis.
      ;
      • Proksch E.
      • Brandner J.M.
      • Jensen J.M.
      The skin: an indispensable barrier.
      ). Melanocytes originate from neural crest cells, migrate to the basement membrane (as observed for both human and mouse skin), and establish contact with keratinocytes (
      • Fuchs E.
      • Raghavan S.
      Getting under the skin of epidermal morphogenesis.
      ;
      • Proksch E.
      • Brandner J.M.
      • Jensen J.M.
      The skin: an indispensable barrier.
      ). Migration of melanocytes to the epidermis and hair follicles accounts for pigmentation in humans (
      • Hearing V.J.
      Biochemical control of melanogenesis and melanosomal organization.
      ;
      • Haass N.K.
      • Herlyn M.
      Normal human melanocyte homeostasis as a paradigm for understanding melanoma.
      ). In mouse neonatal skin, epidermal melanocytes are usually found in the early weeks after birth, whereas in adult skin they are absent from epidermis and reside only in hair follicles and dermis (
      • Quevedo Jr, W.C.
      • Fleischmann R.D.
      Developmental biology of mammalian melanocytes.
      ;
      • Hirobe T.
      Histochemical survey of the distribution of the epidermal melanoblasts and melanocytes in the mouse during fetal and postnatal periods.
      ,
      • Hirobe T.
      Role of keratinocyte-derived factors involved in regulating the proliferation and differentiation of mammalian epidermal melanocytes.
      ).
      Exposure to UVR is one of the major causes of skin cancer (
      • Strom S.S.
      • Yamamura Y.
      Epidemiology of nonmelanoma skin cancer.
      ;
      • Saraiya M.
      • Glanz K.
      • Briss P.A.
      • et al.
      Interventions to prevent skin cancer by reducing exposure to ultraviolet radiation: a systematic review.
      ). Epidemiological studies have suggested that childhood sunburn poses a significant risk of developing aggressive melanoma (
      • Holman C.D.
      • Armstrong B.K.
      • Heenan P.J.
      A theory of the etiology and pathogenesis of human cutaneous malignant melanoma.
      ;
      • Whiteman D.C.
      • Whiteman C.A.
      • Green A.C.
      Childhood sun exposure as a risk factor for melanoma: a systematic review of epidemiologic studies.
      ). UVB radiation (280–320nm) can be absorbed by skin, causing sunburn, epidermal hyperplasia, skin aging, immune suppression, and eventually skin cancer (
      • Young A.R.
      Cumulative effects of ultraviolet radiation on the skin: cancer and photoaging.
      ;
      • Beissert S.
      • Schwarz T.
      Mechanisms involved in ultraviolet light-induced immunosuppression.
      ;
      • Bowden G.T.
      Prevention of non-melanoma skin cancer by targeting ultraviolet-B-light signalling.
      ). UVA (320–400nm) is involved in skin carcinogenesis through indirect DNA damage, such as free radical formation and oxidative reactions (
      • Yasui H.
      • Sakurai H.
      Chemiluminescent detection and imaging of reactive oxygen species in live mouse skin exposed to UVA.
      ;
      • Bickers D.R.
      • Athar M.
      Oxidative stress in the pathogenesis of skin disease.
      ). Previous studies have demonstrated that increased susceptibility of newborn, but not adult, mice is linked with melanoma in later life (
      • Noonan F.P.
      • Recio J.A.
      • Takayama H.
      • et al.
      Neonatal sunburn and melanoma in mice.
      ;
      • Hacker E.
      • Irwin N.
      • Muller H.K.
      • et al.
      Neonatal ultraviolet radiation exposure is critical for malignant melanoma induction in pigmented Tpras transgenic mice.
      ). Moreover, a single erythemal dose of ∼900mJcm-2 (9kJm-2) to 2- to 4-day-old neonates was shown to be more effective than chronic treatments at inducing melanoma in mouse models (
      • Noonan F.P.
      • Recio J.A.
      • Takayama H.
      • et al.
      Neonatal sunburn and melanoma in mice.
      ;
      • Hacker E.
      • Irwin N.
      • Muller H.K.
      • et al.
      Neonatal ultraviolet radiation exposure is critical for malignant melanoma induction in pigmented Tpras transgenic mice.
      ,
      • Hacker E.
      • Muller H.K.
      • Irwin N.
      • et al.
      Spontaneous and UV radiation-induced multiple metastatic melanomas in Cdk4R24C/R24C/TPras mice.
      ;
      • Wolnicka-Glubisz A.
      • Noonan F.P.
      Neonatal susceptibility to UV induced cutaneous malignant melanoma in a mouse model.
      ).
      The first step in UV-induced skin carcinogenesis is DNA damage in the skin, which includes mainly cyclobutane pyrimidine dimers (CPD) and pyrimidine (6–4) pyrimidone photoproducts (6–4 PP) (
      • Setlow R.B.
      • Swenson P.A.
      • Carrier W.L.
      Thymine dimers and inhibition of DNA synthesis by ultraviolet irradiation of cells.
      ;
      • Mitchell D.L.
      • Nairn R.S.
      The biology of the (6-4) photoproduct.
      ;
      • Friedberg E.C.
      DNA damage and repair.
      ).UVR increases oxidative stress-induced formation of 8-hydroxy-2′-deoxyguanosine (8-oxo-dG), which is also linked to skin carcinogenesis (
      • Hattori Y.
      • Nishigori C.
      • Tanaka T.
      • et al.
      8-Hydroxy-2′-deoxyguanosine is increased in epidermal cells of hairless mice after chronic ultraviolet B exposure.
      ;
      • Wilgus T.A.
      • Parrett M.L.
      • Ross M.S.
      • et al.
      Inhibition of ultraviolet light B-induced cutaneous inflammation by a specific cyclooxygenase-2 inhibitor.
      ;
      • Kunisada M.
      • Sakumi K.
      • Tominaga Y.
      • et al.
      8-Oxoguanine formation induced by chronic UVB exposure makes Ogg1 knockout mice susceptible to skin carcinogenesis.
      ;
      • Wulff B.C.
      • Schick J.S.
      • Thomas-Ahner J.M.
      • et al.
      Topical treatment with OGG1 enzyme affects UVB-induced skin carcinogenesis.
      ). The next steps include cell cycle arrest, DNA repair, immunosuppression, gene mutation, and transformation (
      • Ouhtit A.
      • Muller H.K.
      • Davis D.W.
      • et al.
      Temporal events in skin injury and the early adaptive responses in ultraviolet-irradiated mouse skin.
      ;
      • Olivier M.
      • Hussain S.P.
      • Caron de Fromentel C.
      • et al.
      TP53 mutation spectra and load: a tool for generating hypotheses on the etiology of cancer.
      ;
      • D’Errico M.
      • Lemma T.
      • Calcagnile A.
      • et al.
      Cell type and DNA damage specific response of human skin cells to environmental agents.
      ;
      • de Gruijl F.R.
      UV-induced immunosuppression in the balance.
      ). In general, keratinocytes give rise to daughter cells that terminally differentiate; however, if the DNA damage cannot be repaired during the period of cell cycle arrest, keratinocytes are prone to die via apoptosis (
      • Eckert R.L.
      • Efimova T.
      • Dashti S.R.
      • et al.
      Keratinocyte survival, differentiation, and death: many roads lead to mitogen-activated protein kinase.
      ). p53 is induced in both keratinocytes and fibroblasts after UVB irradiation and has a unique role in inducing apoptotic pathways (
      • D’Errico M.
      • Lemma T.
      • Calcagnile A.
      • et al.
      Cell type and DNA damage specific response of human skin cells to environmental agents.
      ;
      • Donehower L.A.
      • Lozano G.
      20 years studying p53 functions in genetically engineered mice.
      ). UVR also stimulates expression of keratinocyte-derived autocrine factors, such as cyclooxygenase-2 (COX-2), GM-CSF, and ILs in mouse skin (
      • Tripp C.S.
      • Blomme E.A.
      • Chinn K.S.
      • et al.
      Epidermal COX-2 induction following ultraviolet irradiation: suggested mechanism for the role of COX-2 inhibition in photoprotection.
      ;
      • Hirobe T.
      • Furuya R.
      • Hara E.
      • et al.
      Granulocyte-macrophage colony-stimulating factor (GM-CSF) controls the proliferation and differentiation of mouse epidermal melanocytes from pigmented spots induced by ultraviolet radiation B.
      ,
      • Hirobe T.
      • Furuya R.
      • Ifuku O.
      • et al.
      Granulocyte-macrophage colony-stimulating factor is a keratinocyte-derived factor involved in regulating the proliferation and differentiation of neonatal mouse epidermal melanocytes in culture.
      ;
      • Rundhaug J.E.
      • Mikulec C.
      • Pavone A.
      • et al.
      A role for cyclooxygenase-2 in ultraviolet light-induced skin carcinogenesis.
      ). Paracrine factors, such as endothelins and pro-opiomelanocortin (POMC), which are secreted by keratinocytes and regulate melanocyte proliferation, have been shown to be altered by UVR exposure (
      • Imokawa G.
      • Yada Y.
      • Miyagishi M.
      Endothelins secreted from human keratinocytes are intrinsic mitogens for human melanocytes.
      ;
      • Chakraborty A.K.
      • Funasaka Y.
      • Slominski A.
      • et al.
      Production and release of proopiomelanocortin (POMC) derived peptides by human melanocytes and keratinocytes in culture: regulation by ultraviolet B.
      ;
      • Kadekaro A.L.
      • Abdel-Malek Z.A.
      Walking in the footsteps of giants: melanocortins and human pigmentation, a historical perspective.
      ).
      Retinoids have been shown to regulate skin development, differentiation, and homeostasis, which are mediated by nuclear receptors such as retinoid acid receptors and retinoid X receptors (RXRs) (
      • Chambon P.
      The molecular and genetic dissection of the retinoid signalling pathway.
      ,
      • Chambon P.
      A decade of molecular biology of retinoic acid receptors.
      ). RXRα, the most abundant isoform of RXR in skin, is a central transcriptional regulator in modulating gene expression by heterodimerization with other nuclear receptors (
      • Li M.
      • Chiba H.
      • Warot X.
      • et al.
      RXR-alpha ablation in skin keratinocytes results in alopecia and epidermal alterations.
      ). RXRα is expressed in both mouse and human epidermis, and RXRα null mice are embryonic lethal between E13.5 and E16.5 (
      • Kastner P.
      • Grondona J.M.
      • Mark M.
      • et al.
      Genetic analysis of RXR alpha developmental function: convergence of RXR and RAR signaling pathways in heart and eye morphogenesis.
      ;
      • Sucov H.M.
      • Dyson E.
      • Gumeringer C.L.
      • et al.
      RXR alpha mutant mice establish a genetic basis for vitamin A signaling in heart morphogenesis.
      ). Using a Cre-loxP strategy, we were able to obtain RXRαep-/- mice that selectively lacked RXRα in epidermal keratinocytes (
      • Li M.
      • Indra A.K.
      • Warot X.
      • et al.
      Skin abnormalities generated by temporally controlled RXRalpha mutations in mouse epidermis.
      ;
      • Metzger D.
      • Indra A.K.
      • Li M.
      • et al.
      Targeted conditional somatic mutagenesis in the mouse: temporally-controlled knock out of retinoid receptors in epidermal keratinocytes.
      ).
      Previous studies have reported that RXRα ablation in skin caused alopecia and altered homeostasis of proliferation/differentiation of epidermal keratinocytes (
      • Li M.
      • Chiba H.
      • Warot X.
      • et al.
      RXR-alpha ablation in skin keratinocytes results in alopecia and epidermal alterations.
      ). RXRα also has a protective role in DMBA/TPA-induced formation of epidermal and melanocytic tumors (
      • Indra A.K.
      • Castaneda E.
      • Antal M.C.
      • et al.
      Malignant transformation of DMBA/TPA-induced papillomas and nevi in the skin of mice selectively lacking retinoid-X-receptor alpha in epidermal keratinocytes.
      ;
      • Hyter S.
      • Bajaj G.
      • Liang X.
      • et al.
      Loss of nuclear receptor RXRalpha in epidermal keratinocytes promotes the formation of Cdk4-activated invasive melanomas [Internet].
      ). To investigate the possible role of keratinocytic RXRα in UV-induced skin damage, we subjected RXRαep-/- and control (CT) RXRαL2/L2 neonatal mice to an acute UVR to study keratinocyte and melanocyte homeostasis. UVR caused increased apoptosis, proliferation, and differentiation in keratinocytes of RXRαep-/- compared with CT mice. UVR also enhanced oxidative stress-induced DNA damage in melanocytes, as well as increasing melanocyte proliferation in both epidermis and dermis of RXRαep-/- mice. Our study data indicated a critical role of keratinocytic RXRα in maintaining skin homeostasis after UV exposure.

      Results

      UVR enhances formation of DNA adducts in RXRαep-/- mice

      To evaluate DNA damage level after UVR, we exposed dorsal skin of CT and RXRαep-/- neonatal mice to a single dose of UVR (UVA and UVB) and collected it after 24, 48, and 72hours (see Materials and Methods). Without UV treatment, very few thymine dimers were observed in CT and RXRαep-/- mouse epidermis. At 24 and 48hours after UV exposure, the number of thymine dimer-positive keratinocytes and melanocytes significantly increased and then dropped at 72hours (Figure 1a; Supplementary Figure S1 online). However, CPD formation in epidermis was not influenced by RXRα ablation (Figure 1a; data not shown). Similarly, oxidative stress-induced DNA damage in both epidermis and dermis was determined by immunohistochemistry (IHC) using anti-8-oxo-dG and anti-TRP2 antibodies. At 48hours after UV exposure, the numbers of 8-oxo-dG-positive cells were significantly increased in both control and mutant mice, and the numbers dropped at 72hours. Most of the oxidative DNA damage occurred in dermal melanocytes, and ∼2-fold more positive cells were found in RXRαep-/- mice compared with CT mice (Figure 1b). The above results suggest that RXRα has a protective effect on oxidative stress-induced DNA damage in mouse skin.
      Figure thumbnail gr1
      Figure 1UVR-induced DNA damage and apoptotic responses in control (CT) and RXRαep-/- neonatal mice skin. (a) Ex vivo immunohistochemical analysis of DNA fragmentation in dorsal skin by detecting formation of thymine dimers (red). (b) Immunohistochemical detection of 8-oxo-dG (red) and TRP2 (green)-positive cells in skin. Yellow (merged panel) indicates Trp2+/8-oxo-dG+ cells. (c) TUNEL assay for detection of apoptotic cells in nonirradiated and UV-irradiated CT and RXRαep-/- skin. Red arrows indicate TUNEL-positive cells. (d) Percentage TUNEL+ (green) cells in CT and RXRαep-/- mice epidermis before and after UVR. Percentage positive cells were counted as a percentage of total number of DAPI+ cells (blue) in epidermis. D, dermis; E, epidermis; RXRα, retinoid X receptor-α. Bar=15.6μm. Statistical analyses were performed by unpaired t-test; **P<0.005; #no statistical difference.

      Increased apoptosis in RXRαep-/- mice after UV irradiation

      TUNEL assays are used to detect DNA strand breaks, which are often associated with apoptosis (
      • Lu Y.P.
      • Lou Y.R.
      • Yen P.
      • et al.
      Time course for early adaptive responses to ultraviolet B light in the epidermis of SKH-1 mice.
      ;
      • Ouhtit A.
      • Muller H.K.
      • Davis D.W.
      • et al.
      Temporal events in skin injury and the early adaptive responses in ultraviolet-irradiated mouse skin.
      ;
      • DeCoster M.A.
      Group III secreted phospholipase A2 causes apoptosis in rat primary cortical neuronal cultures.
      ;
      • Matsumura Y.
      • Moodycliffe A.M.
      • Nghiem D.X.
      • et al.
      Resistance of CD1d-/- mice to ultraviolet-induced skin cancer is associated with increased apoptosis.
      ). The percentage of epidermal TUNEL-positive cells in RXRαep-/- mice was three- and twofold higher at 24 and 72hours, respectively, compared with the controls (Figure 1c and d). At 48hours, although the percentage of TUNEL-positive cells was similar in epidermis, the apoptosis rate was higher in RXRαep-/- mice compared with CT mice in the dermis, indicating a protective role of keratinocytic RXRα in reducing strand-break formation in the dermal compartment (Figure 1c). Similar results were obtained by IHC using an antibody against caspase-3, which is a mediator of the execution phase in apoptosis (data not shown;
      • Cohen G.M.
      Caspases: the executioners of apoptosis.
      ). No difference in apoptosis was observed between nonirradiated CT and RXRαep-/- mice at all time points (see Supplementary Figure S3a online). Together, the above results suggest that keratinocytic RXRα has a role in regulating apoptosis in both epidermis and dermis.

      RXRαep-/- mice exhibited altered epidermal proliferation and differentiation after UV irradiation

      Histological analysis of hematoxylin and eosin–stained skin sections revealed the presence of significant epidermal hyperplasia in RXRαep-/- mice compared with CT mice 48 and 72hours after UVR (Figure 2a). To determine the changes in keratinocyte proliferation of RXRαep-/- mice, we performed IHC analyses for proliferation markers Ki67 and cytokeratin 14 (K14) (see Figure 2b and Supplementary Figure S4a online). The number of epidermal Ki67-positive cells was twofold higher in RXRαep-/- mice 24 and 48hours after acute UV exposure (see Figure 2b and d). Similarly, western blot analyses revealed that expression of K14 peaked 48hours after UVR and was higher in the mutant skin compared with CT skin before and 24 and 72hours after UVR (Figure 2e). No significant difference in percentage of Ki67-positive cells was found between nonirradiated CT and RXRαep-/- mouse skin at all time points (Supplementary Figure S3b online). The expression levels of early (keratin 10, K10) and late (filaggrin and loricrin) differentiation markers were induced after UVR in both CT and RXRαep-/- skin and were higher in mutants compared with CT skin at all time points (24, 48, and 72hours) after UV irradiation (Supplementary Figure S4b–d online). These studies indicate that keratinocytic RXRα has a protective role in skin keratinocytes after UV irradiation and that selective ablation of RXRα in epidermal keratinocytes leads to increased keratinocyte proliferation and differentiation.
      Figure thumbnail gr2
      Figure 2Enhanced keratinocyte proliferative responses in CT and RXRαep-/- mice on UV stimulation. (a) Hematoxylin and eosin-stained 5μm thick paraffin sections from dorsal skin of neonatal CT and RXRαep-/- mice. Bar=31.2μm. (b) Immunohistochemical staining with anti-Ki67 antibody (red). Bar=15.6μm. (c) Measurement of epidermal thickness in CT and RXRαep-/- skin. (d) Percentage of Ki67-positive keratinocytes in CT and RXRαep-/- skin. (e) Western blot analysis of K14 expression in CT and RXRαep-/- mice before and after UV irradiation. β-Actin was internal control. Separations between blots indicate lanes not originally run next to each other and juxtaposed in the figure. Sections were counterstained with DAPI (blue). Statistical analyses were performed by unpaired t-test and corrected with Bonferroni step-down analysis; *P<0.05; **P<0.005; #no statistical difference. CT, control; D, dermis; E, epidermis; HF, hair follicles; RXRα, retinoid X receptor-α.

      Ablation of RXRα in epidermis increased expression of keratinocyte-derived autocrine factors

      To determine whether altered keratinocyte proliferation and differentiation in mutant skin are due to impaired mitogenic responses, we performed quantitative RT-PCR (qRT-PCR) analyses for diffusible growth factors such as heparin-binding EGF-like growth factor and GM-CSF on RNA isolated from control and mutant mice (
      • Braunstein S.
      • Kaplan G.
      • Gottlieb A.B.
      • et al.
      GM-CSF activates regenerative epidermal growth and stimulates keratinocyte proliferation in human skin in vivo.
      ;
      • Werner S.
      • Smola H.
      Paracrine regulation of keratinocyte proliferation and differentiation.
      ;
      • Yoshida A.
      • Kanno H.
      • Watabe D.
      • et al.
      The role of heparin-binding EGF-like growth factor and amphiregulin in the epidermal proliferation of psoriasis in cooperation with TNFalpha.
      ). Expression of both factors was upregulated twofold in skin of RXRαep-/- compared with CT mice and was highest 24hours after UVR (Figure 3a). Relative expression of IL-1α, a major proinflammatory cytokine, was modestly increased in the mutant mice before and 24hours after UVR (see Figure 3b). Similarly, relative expression level of COX-2, which is important in inflammation, was significantly higher in the mutant skin 24hours after UV treatment (see Figure 3b). Expression of all these genes dropped 72hours after UV except for COX-2. These results showed that keratinocytic RXRα, in a cell-autonomous manner, modulates keratinocyte proliferation and/or differentiation via secretion of mitogenic growth factors and cytokines within a short time window after UVR.
      Figure thumbnail gr3
      Figure 3Quantitative RT-PCR analysis of expression of paracrine and autocrine factors, and western blot analysis of protein expression in skin of CT and RXRαep-/- (MT) mice. Relative expression levels of (a) hbEGF, GM-CSF; (b) COX-2, IL-1α; (c) ET-1, FGF2; and (d) POMC, SCF at different time points after UVR, are indicated. Values represent relative transcript level after normalization with HPRT transcripts. (e) Expression levels of different proteins such as ERK (p42/44), phospho-ERK, JNK, phospho-JNK, p53, and p21 were analyzed by western blot in control and mutant skin extracts obtained before and after UVR. β-Actin was used as an internal control. Actual band sizes are indicated on the left of panel. *P<0.2; **P<0.05; #no statistical difference between CT and RXRαep-/- mice. COX-2, cyclooxygenase-2; CT, control; RXRα, retinoid X receptor-α.

      Altered mitogenic and apoptotic responses in the skin of RXRαep-/- mice after UVR

      We hypothesized that increased proliferation and differentiation and altered apoptosis in RXRαep-/- mice may be due to alteration of mitogenic and apoptotic signaling following UV irradiation. To test this, we performed western blot analyses for determining the protein levels and phosphorylation status of key mitogen-activated protein kinase proteins (ERK and JNK) and the expression level of transcription factor p53 in CT and RXRαep-/- (MT) mouse skin (see Figure 3e;
      • Eckert R.L.
      • Efimova T.
      • Dashti S.R.
      • et al.
      Keratinocyte survival, differentiation, and death: many roads lead to mitogen-activated protein kinase.
      ;
      • Kulesz-Martin M.
      • Lagowski J.
      • Fei S.
      • et al.
      Melanocyte and keratinocyte carcinogenesis: p53 family protein activities and intersecting mRNA expression profiles.
      ). The expression levels of both ERK and JNK were similar between control and mutant skin at 0 and 24hours time points and modestly higher in mutant at 72hours. Interestingly, the protein levels of phospho-ERK and phospho-JNK were slightly lower at 24hours, but higher at 72hours in mutant, compared with the CT skin (Figure 3e). Expression of tumor suppressor p53 and its direct target p21 was similarly induced in both control and mutant skin after UVR. Induction of both p53 and p21 peaked after 24hours of UVR in CT mice. However, in mutant mouse skin, these induction levels peaked 72hours after UV irradiation, suggesting delayed p53-mediated cellular responses in RXRαep-/- mice. The above results suggest that enhanced proliferation and differentiation and increased apoptosis in the mutant skin can be due to aberrant growth-promoting mitogenic and altered apoptotic responses in these mice following UV irradiation.

      Keratinocytic RXRα is involved in keratinocyte–melanocyte cross talk

      It is possible that ablation of RXRα in epidermal keratinocytes impacts melanocyte homeostasis. We therefore performed double staining with antibodies against proliferation marker PCNA and melanocyte-specific marker TRP1 on control and mutant mouse skin. Without UVR, only a few melanocytes were identified in CT and RXRαep-/- mouse epidermis 2 days postnatal (P2) (Figure 4a, left panel; Figure 4c). At 48hours after UVR, the number of epidermal melanocytes in CT mice increased ∼twofold (∼two per field) as compared with nonirradiated skin. In contrast, epidermal melanocytes in mutant were significantly higher as compared with CT (∼three per field) at 48hours. At 72hours after UVR, epidermal melanocyte number peaked for both control (∼four per field) and mutant mice (∼4.5 per field) (Figure 4a and c). We also performed Fontana–Masson staining to detect pigmented melanocytes in CT and RXRαep-/- skin at all time points (Figure 4b; data not shown). Comparable amounts of melanocytes were found in the dermis of control and mutant mice before and 24hours after UVR (Figure 4b). At 24hours, large clusters of pigmented melanocytes were found in the dermis (Figure 4b, middle panel). Interestingly, 48hours after UVR, RXRαep-/- mouse skin exhibited ∼twofold more pigmented melanocytes in the dermal compartment compared with CT (Figure 4b, right panel; Figure 4d). Consistently, increased TRP2-positive melanocytes were detected in the dermis of mutant skin 48hours after UVR (see Figure 1b; data not shown). The above results suggest that RXRαep-/- mice are more sensitive to UVR-induced melanocyte activation/proliferation in both epidermis and dermis.
      Figure thumbnail gr4
      Figure 4Effects of keratinocytic RXRα on melanocyte proliferation after UV exposure. (a) IHC staining for proliferating melanocytes with PCNA (green) and TRP1 (red). Nuclei were labeled with DAPI (blue). Arrows indicate epidermal melanocytes. (b) Fontana–Masson staining of control and RXRαep-/- skin. Nuclei (pink-red), cytoplasm (pink), and melanin-producing melanocytes (black) are indicated. Arrows indicate melanin-producing melanocytes in dermis. Bar=15.6μm. Counts of melanocytes in CT and RXRαep-/- (c) epidermis and (d) dermis at different time points. (e) Proliferation assay of primary melanocytes 24hours after culture. Statistical analyses were performed by two-tailed unpaired t-test and corrected with Bonferroni step-down analysis using GraphPad Prism software; *P<0.1; **P<0.05; #no statistical difference. CT, control; IHC, immunohistochemistry; KM, keratinocyte medium; RXRα, retinoid X receptor-α.
      To confirm the ex vivo data, we cultured primary melanocytes in keratinocyte-derived conditioned medium from CT and mutant mice. Melanocyte growth medium was used as control. Murine melanocytes showed an increased growth 24hours after incubation with keratinocyte-conditioned medium from control or mutant mice compared with their own medium (see Figure 4e). Melanocytes cultured 24hours in keratinocyte-derived conditioned medium from RXRαep-/- mouse skin after UVR had higher proliferation than those cultured in conditioned medium from CT skin (Figure 4e).
      To determine the mechanism(s) of keratinocytic RXRα-modulated melanocyte proliferation, we studied the relative mRNA expression levels of paracrine factors such as endothelin 1 (ET-1), fibroblast growth factor-2 (FGF2), stem cell factor (SCF), and POMC, which are secreted by keratinocytes to stimulate melanocyte proliferation (Figure 4c and d;
      • Yada Y.
      • Higuchi K.
      • Imokawa G.
      Effects of endothelins on signal transduction and proliferation in human melanocytes.
      ;
      • Imokawa G.
      • Yada Y.
      • Miyagishi M.
      Endothelins secreted from human keratinocytes are intrinsic mitogens for human melanocytes.
      ;
      • Slominski A.
      • Paus R.
      Melanogenesis is coupled to murine anagen: toward new concepts for the role of melanocytes and the regulation of melanogenesis in hair growth.
      ;
      • Chakraborty A.K.
      • Funasaka Y.
      • Slominski A.
      • et al.
      Production and release of proopiomelanocortin (POMC) derived peptides by human melanocytes and keratinocytes in culture: regulation by ultraviolet B.
      ;
      • Haass N.K.
      • Herlyn M.
      Normal human melanocyte homeostasis as a paradigm for understanding melanoma.
      ). Relative mRNA expression levels of ET-1, FGF2, and SCF were more than twofold higher in mutant skin before and 24hours after UV irradiation (Figure 3c and d). Overall, the results of our study provide compelling evidence that keratinocytic RXRα, in cooperation with UVR, regulates melanocyte homeostasis by repressing secretion of paracrine mitogenic growth factors.

      Discussion

      We have identified RXRα as a key regulator of UVR-induced cellular responses. Ablation of RXRα in neonatal mouse epidermis causes skin hyperplasia, which is reflected by increased epidermal proliferation and differentiation. RXRα also regulated oxidative stress-induced DNA damage and skin apoptosis in cell-autonomous and non-cell-autonomous manners. Moreover, epidermal RXRα was found to be involved in suppressing activation and/or proliferation of epidermal and dermal melanocytes after acute UV irradiation, mainly by controlling the keratinocyte–melanocyte cross talk through secretion of paracrine factors.

      Keratinocytic RXRα protects melanocytes from oxidative stress–induced DNA damage after UVR

      We have found that CPD formation in keratinocytes begins early after UVR and lasts for at least 2 days. In contrast, CPD formation in melanocytes was observed approximately 48hours after UVR and decreased soon afterward, indicating the possibility of a different mechanism of DNA repair in melanocytes relative to keratinocytes. Similar results of UVR-induced CPD formation in murine keratinocytes and melanocytes have previously been reported (
      • Walker G.J.
      • Kimlin M.G.
      • Hacker E.
      • et al.
      Murine neonatal melanocytes exhibit a heightened proliferative response to ultraviolet radiation and migrate to the epidermal basal layer.
      ). The present study has shown that dermal melanocytes in the skin of mutant mice were more susceptible to oxidative stress–induced DNA damage as confirmed by increased positive staining for 8-oxo-dG (Figure 1b), the mechanism of which needs to be better understood. In human melanocytes, UV induces reactive oxygen species that cause 8-oxo-dG signature mutation after UVR (
      • Charron R.A.
      • Fenwick J.C.
      • Lean D.R.
      • et al.
      Ultraviolet-B radiation effects on antioxidant status and survival in the zebrafish, Brachydanio rerio.
      ;
      • Jin G.H.
      • Liu Y.
      • Jin S.Z.
      • et al.
      UVB induced oxidative stress in human keratinocytes and protective effect of antioxidant agents.
      ).
      Paracrine factor α-melanocyte-stimulating hormone has been shown to protect human melanocytes against UV-induced oxidative stress (
      • Song X.
      • Mosby N.
      • Yang J.
      • et al.
      alpha-MSH activates immediate defense responses to UV-induced oxidative stress in human melanocytes.
      ). Here we observed an increased expression of POMC (precursor of α-melanocyte stimulating hormone) and also an increase in oxidative stress in the mutant skin, suggesting an alternative mechanism of keratinocytic RXRα-mediated protection of melanocytes against reactive oxygen species–induced oxidative stress after UVR. To the best of our knowledge, these results have not previously been reported. The differences between the previous in vitro study and our ex vivo study can be attributed to (i) the inherent differences between human and mouse melanocytes, (ii) the epidermal versus dermal origin of melanocytes in the two studies, and/or (iii) complex microenvironmental effects in an ex vivo study compared with an in vitro study. Overall, these results demonstrate that keratinocytic RXRα, in a non-cell-autonomous manner, protects melanocytes and other dermal cells from oxidative stress–induced DNA damage following solar UV irradiation.

      Altered UVR-induced apoptotic responses in mutant skin

      In neonatal mouse skin, we observed that epidermal keratinocyte apoptosis significantly increased and peaked at 24hours after UVR, which then dropped at around 72hours, thereby corroborating results reported earlier for adult mice by
      • Ouhtit A.
      • Muller H.K.
      • Davis D.W.
      • et al.
      Temporal events in skin injury and the early adaptive responses in ultraviolet-irradiated mouse skin.
      . The percentage of apoptotic cells was higher in RXRαep-/- skin than in CT mice at all time points after UV exposure, suggesting a higher penetration of UVR through the epidermal barrier to the dermis in RXRαep-/- mice. The increased UVR-induced apoptosis in the mutant skin can be attributed to additional DNA damage besides thymine dimer formation, impaired DNA repair mechanism, and/or altered apoptotic responses.
      UVB induces p53 expression in wild-type adult mouse skin with a maximum level at 24hours after a single dose of UVR (
      • Campbell C.
      • Quinn A.G.
      • Angus B.
      • et al.
      Wavelength specific patterns of p53 induction in human skin following exposure to UV radiation.
      ;
      • Matsumura Y.
      • Moodycliffe A.M.
      • Nghiem D.X.
      • et al.
      Resistance of CD1d-/- mice to ultraviolet-induced skin cancer is associated with increased apoptosis.
      ). The results of our western blot analysis and p53 IHC staining confirmed that RXRα ablation in epidermis delayed the induction of p53 and, in turn, its target gene p21 expression beyond 24hours, which did not explain the increased apoptosis in RXRαep-/- mice 24hours after UV exposure (Figure 3; Supplementary Figure S2 online). These results suggest that increased apoptotic responses in RXRαep-/- mice may be mediated by both p53-dependent and -independent pathways.

      Keratinocytic RXRα regulates UVR-induced melanocyte proliferation/activation via secretion of paracrine factors

      Melanocytes were shown to respond to UVB exposure and exhibited an increase in proliferation in both human and mouse skin compared with no-treatment groups (
      • Sato T.
      • Kawada A.
      Mitotic activity of hairless mouse epidermal melanocytes: its role in the increase of melanocytes during ultraviolet radiation.
      ;
      • Stierner U.
      • Rosdahl I.
      • Augustsson A.
      • et al.
      UVB irradiation induces melanocyte increase in both exposed and shielded human skin.
      ;
      • van Schanke A.
      • Jongsma M.J.
      • Bisschop R.
      • et al.
      Single UVB overexposure stimulates melanocyte proliferation in murine skin, in contrast to fractionated or UVA-1 exposure.
      ;
      • Walker G.J.
      • Kimlin M.G.
      • Hacker E.
      • et al.
      Murine neonatal melanocytes exhibit a heightened proliferative response to ultraviolet radiation and migrate to the epidermal basal layer.
      ). Our data of TRP1 staining on neonatal CT and mutant mice confirmed that epidermal melanocyte numbers peaked 3 days after UVR, which corroborated well with an earlier report (
      • Walker G.J.
      • Kimlin M.G.
      • Hacker E.
      • et al.
      Murine neonatal melanocytes exhibit a heightened proliferative response to ultraviolet radiation and migrate to the epidermal basal layer.
      ). However, in RXRαep-/- mice, the numbers of epidermal and dermal melanocytes were always higher after UVR, suggesting a rapid response to UV effects in those mice, which may be due to increased sensitivity and susceptibility of RXRαep-/- mice to UVR (see Figure 4a, c and d). Our in vitro proliferation assay showed a significant increase in melanocytes after culture with conditioned keratinocyte medium from RXRαep-/- mice, compared with the control conditioned medium, following UV exposure; this indicates an enhanced secretion of growth-promoting mitogenic paracrine factors from mutant keratinocytes. qRT-PCR studies also confirmed increased expression of several mitogenic paracrine factors such as ET-1, POMC, SCF, and FGF2 in the mutant skin, all of which have been reported to regulate melanocyte proliferation and melanogenesis (
      • Yada Y.
      • Higuchi K.
      • Imokawa G.
      Effects of endothelins on signal transduction and proliferation in human melanocytes.
      ;
      • Imokawa G.
      • Yada Y.
      • Miyagishi M.
      Endothelins secreted from human keratinocytes are intrinsic mitogens for human melanocytes.
      ;
      • Slominski A.
      • Paus R.
      Melanogenesis is coupled to murine anagen: toward new concepts for the role of melanocytes and the regulation of melanogenesis in hair growth.
      ;
      • Chakraborty A.K.
      • Funasaka Y.
      • Slominski A.
      • et al.
      Production and release of proopiomelanocortin (POMC) derived peptides by human melanocytes and keratinocytes in culture: regulation by ultraviolet B.
      ). Recent studies suggested that p53 promotes melanogenic cytokine POMC expression, as well as ET-1 and SCF production, in epidermal hyperpigmented keratinocytes induced by UVR (
      • Cui R.
      • Widlund H.R.
      • Feige E.
      • et al.
      Central role of p53 in the suntan response and pathologic hyperpigmentation.
      ;
      • Murase D.
      • Hachiya A.
      • Amano Y.
      • et al.
      The essential role of p53 in hyperpigmentation of the skin via regulation of paracrine melanogenic cytokine receptor signaling.
      ). Chromatin immunoprecipitation assay on mouse skin extracts showed no direct recruitment of RXRα on the p53 promoter (
      • Hyter S.
      • Bajaj G.
      • Liang X.
      • et al.
      Loss of nuclear receptor RXRalpha in epidermal keratinocytes promotes the formation of Cdk4-activated invasive melanomas [Internet].
      ). Therefore, the mechanism of deregulated p53 expression in the mutant mice following UV exposure is currently unknown. These results suggest multiple levels of gene regulation and confirm that RXRα can directly or indirectly modulate expression of genes encoding paracrine factors involved in modulating melanocyte homeostasis.
      We have found that RXRα modulates skin keratinocyte and melanocyte homeostasis through autocrine and paracrine signaling. Generation of bigenic mice with selective deletion of both RXRα and each one of the paracrine factors will help us reveal the non-cell-autonomous effects mediated by RXRα. RXRα heterodimeric partners that are involved in mediating the UVR-induced cellular responses also need to be identified. Our research is expected to help generate mouse models for studying UV-induced photocarcinogenesis and melanomagenesis and develop efficient strategies to prevent and/or cure UV-induced squamous carcinoma and melanoma.

      Materials and Methods

      Mice

      Mice carrying LoxP-site-containing (floxed) RXRαL2 alleles were bred with hemizygous K14-Cre(tg/0) transgenic mice to produce K14-Cre(tg/0)/RXRαL2/L2 mice (
      • Li M.
      • Indra A.K.
      • Warot X.
      • et al.
      Skin abnormalities generated by temporally controlled RXRalpha mutations in mouse epidermis.
      ;
      • Metzger D.
      • Indra A.K.
      • Li M.
      • et al.
      Targeted conditional somatic mutagenesis in the mouse: temporally-controlled knock out of retinoid receptors in epidermal keratinocytes.
      ). The constitutively active K14-Cre transgene selectively deleted the gene encoding RXRα in epidermis to generate RXRαep-/- mice, and RXRαL2/L2 mice were used as control (
      • Metzger D.
      • Indra A.K.
      • Li M.
      • et al.
      Targeted conditional somatic mutagenesis in the mouse: temporally-controlled knock out of retinoid receptors in epidermal keratinocytes.
      ). RXRα ablation in the epidermis was verified by PCR analyses of DNA isolated from tail biopsies as described (
      • Li M.
      • Indra A.K.
      • Warot X.
      • et al.
      Skin abnormalities generated by temporally controlled RXRalpha mutations in mouse epidermis.
      ). All CT and RXRαep-/- mice were of similar C57BL/6J (∼70%), SV129 (∼20%), and SJL (∼10%) mixed genetic background; 8 to 10 mice from multiple litters were used in each group at each time point. CT and MT littermates were selected in each group and for all time points to minimize variability due to the mixed background. OSU IACUC approval was obtained for animal experiments.

      UVR

      Mice (2 days old, P2) were exposed to a single dose of 900mJcm-2 UV light from a bank of four Philips FS-40 UV sunlamps (New York, NY). Besides the major output of UVB, there was also some output from the UVA spectrum. The irradiance of the sunlamps was measured with an IL-1400A radiometer with an SEE240 UVB detector (International Light, Newburyport, MA).

      Isolation of skin samples

      Mice were euthanized 24, 48, and 72 hours after UV irradiation, and skin samples were retrieved. Mice skin samples for the 0-hour time point were taken 1 day after birth without UV. For each mouse, three pieces of skin (0.5cm2) were taken for IHC, RNA, and protein isolation, respectively.

      Histological analysis

      Skin biopsies were fixed in 4% paraformaldehyde overnight and embedded in paraffin blocks, and 5μm paraffin sections were sectioned using a Leica RM2255 microtome (Bannockburn, IL). Hematoxylin and eosin staining was performed as described elsewhere (
      • Indra A.K.
      • Dupe V.
      • Bornert J.M.
      • et al.
      Temporally controlled targeted somatic mutagenesis in embryonic surface ectoderm and fetal epidermal keratinocytes unveils two distinct developmental functions of BRG1 in limb morphogenesis and skin barrier formation.
      ,
      • Indra A.K.
      • Mohan II, W.S.
      • Frontini M.
      • et al.
      TAF10 is required for the establishment of skin barrier function in foetal, but not in adult mouse epidermis.
      ). Fontana–Masson staining was performed using a commercial kit according to the manufacturer's protocol (American MasterTech, Lodi, CA).

      Immunohistochemistry

      Paraffin sections (5μm) were deparaffinized and rinsed with water, and antigen retrieval was performed with pH 6.0 citrate buffer at 95–100°C for 20minutes. Slides were washed with PBS (3 × ) and blocked with 10% normal goat serum (Vector Laboratories, Burlingame, CA) for 30minutes. Then slides were incubated overnight with primary antibodies (see Supplementary Methods online), followed by three washes with PBS+0.05% Tween 20, and incubated with fluorescently labeled secondary antibody for an hour at room temperature. Nuclei were visualized with 4′,6-diamidino-2-phenyl indole (DAPI). After the final washes, slides were dehydrated through a series of ethanol washes, cleared in xylene, and mounted with DPX mounting media. All images were captured at × 40 magnification using a Carl Zeiss Axio Imager Z1 fluorescent microscope (Thornwood, NY) and AxioCam camera. Data were analyzed and quantified using Adobe Photoshop and ImageJ software (Bethesda, MD). Multiple IHC fields on each slide from control and mutant mouse skin were randomly chosen, and 10–15 such fields per slide were counted. The slides were analyzed independently in a double-blinded manner by two observers. All experiments in each category were repeated at least three times, and the results obtained were very consistent between various experiments.

      Preparation of keratinocyte-conditioned medium

      Dorsal skin taken from neonatal mice was incubated overnight in CnT-07 medium (CELLnTEC, Bern, Switzerland) containing 5mgml-1 dispase (Gibco, Grand Island, NY) and 5 × antibiotic/antimycotic solution (CELLnTEC) at 4°C to dissociate epidermis. Keratinocytes were dissociated from epidermis via incubation in TrypLE Select (Gibco). Cells were collected by centrifugation (160g) and seeded into uncoated 100mm dishes at a density of 5.0 × 104 cells per cm2. Cells were cultured in CnT-07 containing 1 × antibiotic/antimycotic at 35°C, 5% CO2. Growth medium was changed on day 2 of culture. At 48hours after medium change, the medium was either removed or stored at -20°C until use, or culture dishes containing cells were irradiated with 850mJcm-2 UV. Irradiated cells were placed back at 35°C, 5% CO2 for 24hours, after which medium was collected and stored at -20°C until use.

      Melanocyte proliferation assay

      Primary murine melanocytes were obtained from Yale University (New Haven, CT). Cells were maintained in melanocyte growth medium, consisting of OPTI-MEM (Invitrogen, Carlsbad, CA) containing 7% heat-inactivated horse serum (Invitrogen), 10ngml-1 TPA, and 1 × antibiotic/antimycotic (CELLnTEC). Growth conditions were 37°C and 5% CO2. Cells were collected using TrypLE Select (Gibco) and seeded into 96-well white/clear plate (BD Biocoat, San Jose, CA) at a density of 5 × 103 cells per well. Cells were placed under normal growth conditions for 3hours to allow for complete adherence to the plate. Melanocyte growth medium was removed, and cells were placed under starvation conditions overnight using basal DMEM containing 1 × antibiotic/antimycotic. Starvation medium was then removed and replaced with a 1:1 mixture of keratinocyte-conditioned medium (either UV or non-UV treated, RXRαep-/- or RXRαL2/L2 conditioned) and melanocyte growth medium. Unmixed melanocyte growth medium was used as control. Cells were incubated for 24hours at 37°C, 5% CO2, after which medium was removed and cell number assayed using a CyQUANT NF Cell Proliferation Assay Kit (Invitrogen) according to the manufacturer's recommendation. All assays were performed in triplicate.

      qRT-PCR

      RNA extraction and cDNA preparation were performed as described (
      • Indra A.K.
      • Mohan II, W.S.
      • Frontini M.
      • et al.
      TAF10 is required for the establishment of skin barrier function in foetal, but not in adult mouse epidermis.
      ). Real-time PCR was performed on an ABI 7500 Real-Time PCR system using SYBR Green methodology (
      • Indra A.K.
      • Dupe V.
      • Bornert J.M.
      • et al.
      Temporally controlled targeted somatic mutagenesis in embryonic surface ectoderm and fetal epidermal keratinocytes unveils two distinct developmental functions of BRG1 in limb morphogenesis and skin barrier formation.
      ,
      • Indra A.K.
      • Mohan II, W.S.
      • Frontini M.
      • et al.
      TAF10 is required for the establishment of skin barrier function in foetal, but not in adult mouse epidermis.
      ). Relative gene expression analysis of the qRT-PCR data was performed using HPRT as an internal control. The results of our qRT-PCR analysis, using multiple primers from different exons of HPRT, confirmed that its expression did not change as a result of RXRα ablation compared with other controls such as glyceraldehyde-3-phosphate dehydrogenase or β-actin. Primer sequences used are shown in Supplementary Table S1 online. All assays were performed in triplicates.

      Western blotting

      Both control and RXRαep-/- mice were exposed to UV irradiation with a single dose (900mJcm-2), and the skin biopsies were collected at 0, 24, and 72hours after exposure. Skin samples were homogenized, and proteins were extracted with a lysis buffer (150mM NaCl, 50mM Tris (pH 7.5), 5mM EDTA, and 1% Nonidet P-40, 0.1% SDS, 0.5% sodium deoxycholate) containing protease inhibitors (2μgml-1 aprotinin, 2μgml-1 leupeptin, 100μgml-1 Pefabloc, and 1μgml-1 pepstatin). Equal amounts of protein extract (25μg) from each lysate were resolved using SDS-PAGE and transferred onto a nitrocellulose membrane. Blots were blocked overnight with 5% nonfat dry milk and incubated with specific antibodies (Supplementary Methods online). All western blot experiments were carried out in triplicates, using a minimum of three biological replicates from each group of mice.

      Statistical analysis

      Statistical significance of differences between groups was assessed using GraphPad Prism software (La Jolla, CA) by two-tailed unpaired t-test and Bonferroni step-down (Holm) analysis. The TUNEL- or Ki67-positive cells of CT and MT mice were counted; this number was expressed as a percentage of DAPI-positive cells. Similarly, epidermal and dermal melanocytes in multiple skin sections were counted and expressed as melanocytes per field. IHC data were quantified using Adobe Photoshop and ImageJ software, and epidermal thickness was quantified using Leica One-Suite software. Multiple sections were analyzed from 8 to 10 mice of each genotype and for each time point, and significance was determined using Student's unpaired t-test. Data obtained from each group of control and mutant mice for each time point were combined for calculating the mean data and SEM. Sensitivity analysis with mixed models showed that the litter effects are not significant and a simple model with the subset is justified. All statistical analyses were independently performed in a double-blinded manner by two investigators.

      ACKNOWLEDGMENTS

      We thank Pierre Chambon and Daniel Metzger for helpful advice and critically reviewing the article and Cliff Pereira for statistical analysis. We also thank Talicia Savage, Erin Bredeweg, and Steven Ma for help in sample processing. These studies were supported by grant ES016629-01A1 to AI from the National Institute of Environmental and Health Sciences (NIEHS) at the National Institutes of Health, an OHSU Medical Research Foundation grant to AI, and an NIEHS Center grant (ES00210) to the Oregon State University Environmental Health Sciences Center. In addition, we thank Mark Zabriskie, and Gary DeLander of the OSU College of Pharmacy for continuous support and encouragement.

      SUPPLEMENTARY MATERIAL

      Supplementary material is linked to the online version of the paper at http://www.nature.com/jid

      REFERENCES

        • Beissert S.
        • Schwarz T.
        Mechanisms involved in ultraviolet light-induced immunosuppression.
        J Investig Dermatol Symp Proc. 1999; 4: 61-64
        • Bickers D.R.
        • Athar M.
        Oxidative stress in the pathogenesis of skin disease.
        J Invest Dermatol. 2006; 126: 2565-2575
        • Bowden G.T.
        Prevention of non-melanoma skin cancer by targeting ultraviolet-B-light signalling.
        Nat Rev Cancer. 2004; 4: 23-35
        • Braunstein S.
        • Kaplan G.
        • Gottlieb A.B.
        • et al.
        GM-CSF activates regenerative epidermal growth and stimulates keratinocyte proliferation in human skin in vivo.
        J Invest Dermatol. 1994; 103: 601-604
        • Campbell C.
        • Quinn A.G.
        • Angus B.
        • et al.
        Wavelength specific patterns of p53 induction in human skin following exposure to UV radiation.
        Cancer Res. 1993; 53: 2697-2699
        • Chakraborty A.K.
        • Funasaka Y.
        • Slominski A.
        • et al.
        Production and release of proopiomelanocortin (POMC) derived peptides by human melanocytes and keratinocytes in culture: regulation by ultraviolet B.
        Biochim Biophys Acta. 1996; 1313: 130-138
        • Chambon P.
        The molecular and genetic dissection of the retinoid signalling pathway.
        Gene. 1993; 135: 223-228
        • Chambon P.
        A decade of molecular biology of retinoic acid receptors.
        FASEB J. 1996; 10: 940-954
        • Charron R.A.
        • Fenwick J.C.
        • Lean D.R.
        • et al.
        Ultraviolet-B radiation effects on antioxidant status and survival in the zebrafish, Brachydanio rerio.
        Photochem Photobiol. 2000; 72: 327-333
        • Cohen G.M.
        Caspases: the executioners of apoptosis.
        Biochem J. 1997; 326: 1-16
        • Cui R.
        • Widlund H.R.
        • Feige E.
        • et al.
        Central role of p53 in the suntan response and pathologic hyperpigmentation.
        Cell. 2007; 128: 853-864
        • D’Errico M.
        • Lemma T.
        • Calcagnile A.
        • et al.
        Cell type and DNA damage specific response of human skin cells to environmental agents.
        Mutat Res. 2007; 614: 37-47
        • de Gruijl F.R.
        UV-induced immunosuppression in the balance.
        Photochem Photobiol. 2008; 84: 2-9
        • DeCoster M.A.
        Group III secreted phospholipase A2 causes apoptosis in rat primary cortical neuronal cultures.
        Brain Res. 2003; 988: 20-28
        • Donehower L.A.
        • Lozano G.
        20 years studying p53 functions in genetically engineered mice.
        Nat Rev Cancer. 2009; 9: 831-841
        • Eckert R.L.
        • Efimova T.
        • Dashti S.R.
        • et al.
        Keratinocyte survival, differentiation, and death: many roads lead to mitogen-activated protein kinase.
        J Investig Dermatol Symp Proc. 2002; 7: 36-40
        • Friedberg E.C.
        DNA damage and repair.
        Nature. 2003; 421: 436-440
        • Fuchs E.
        • Raghavan S.
        Getting under the skin of epidermal morphogenesis.
        Nat Rev Genet. 2002; 3: 199-209
        • Haass N.K.
        • Herlyn M.
        Normal human melanocyte homeostasis as a paradigm for understanding melanoma.
        J Investig Dermatol Symp Proc. 2005; 10: 153-163
        • Hacker E.
        • Irwin N.
        • Muller H.K.
        • et al.
        Neonatal ultraviolet radiation exposure is critical for malignant melanoma induction in pigmented Tpras transgenic mice.
        J Invest Dermatol. 2005; 125: 1074-1077
        • Hacker E.
        • Muller H.K.
        • Irwin N.
        • et al.
        Spontaneous and UV radiation-induced multiple metastatic melanomas in Cdk4R24C/R24C/TPras mice.
        Cancer Res. 2006; 66: 2946-2952
        • Hattori Y.
        • Nishigori C.
        • Tanaka T.
        • et al.
        8-Hydroxy-2′-deoxyguanosine is increased in epidermal cells of hairless mice after chronic ultraviolet B exposure.
        J Invest Dermatol. 1996; 107: 733-737
        • Hearing V.J.
        Biochemical control of melanogenesis and melanosomal organization.
        J Investig Dermatol Symp Proc. 1999; 4: 24-28
        • Hirobe T.
        Histochemical survey of the distribution of the epidermal melanoblasts and melanocytes in the mouse during fetal and postnatal periods.
        Anat Rec. 1984; 208: 589-594
        • Hirobe T.
        Role of keratinocyte-derived factors involved in regulating the proliferation and differentiation of mammalian epidermal melanocytes.
        Pigment Cell Res. 2005; 18: 2-12
        • Hirobe T.
        • Furuya R.
        • Hara E.
        • et al.
        Granulocyte-macrophage colony-stimulating factor (GM-CSF) controls the proliferation and differentiation of mouse epidermal melanocytes from pigmented spots induced by ultraviolet radiation B.
        Pigment Cell Res. 2004; 17: 230-240
        • Hirobe T.
        • Furuya R.
        • Ifuku O.
        • et al.
        Granulocyte-macrophage colony-stimulating factor is a keratinocyte-derived factor involved in regulating the proliferation and differentiation of neonatal mouse epidermal melanocytes in culture.
        Exp Cell Res. 2004; 297: 593-606
        • Holman C.D.
        • Armstrong B.K.
        • Heenan P.J.
        A theory of the etiology and pathogenesis of human cutaneous malignant melanoma.
        J Natl Cancer Inst. 1983; 71: 651-656
        • Hyter S.
        • Bajaj G.
        • Liang X.
        • et al.
        Loss of nuclear receptor RXRalpha in epidermal keratinocytes promotes the formation of Cdk4-activated invasive melanomas [Internet].
        Pigment Cell Melanoma Res. 2010; 23: 635-648
        • Imokawa G.
        • Yada Y.
        • Miyagishi M.
        Endothelins secreted from human keratinocytes are intrinsic mitogens for human melanocytes.
        J Biol Chem. 1992; 267: 24675-24680
        • Indra A.K.
        • Castaneda E.
        • Antal M.C.
        • et al.
        Malignant transformation of DMBA/TPA-induced papillomas and nevi in the skin of mice selectively lacking retinoid-X-receptor alpha in epidermal keratinocytes.
        J Invest Dermatol. 2007; 127: 1250-1260
        • Indra A.K.
        • Dupe V.
        • Bornert J.M.
        • et al.
        Temporally controlled targeted somatic mutagenesis in embryonic surface ectoderm and fetal epidermal keratinocytes unveils two distinct developmental functions of BRG1 in limb morphogenesis and skin barrier formation.
        Development. 2005; 132: 4533-4544
        • Indra A.K.
        • Mohan II, W.S.
        • Frontini M.
        • et al.
        TAF10 is required for the establishment of skin barrier function in foetal, but not in adult mouse epidermis.
        Dev Biol. 2005; 285: 28-37
        • Jin G.H.
        • Liu Y.
        • Jin S.Z.
        • et al.
        UVB induced oxidative stress in human keratinocytes and protective effect of antioxidant agents.
        Radiat Environ Biophys. 2007; 46: 61-68
        • Kadekaro A.L.
        • Abdel-Malek Z.A.
        Walking in the footsteps of giants: melanocortins and human pigmentation, a historical perspective.
        Pigment Cell Res. 2007; 20: 150-152
        • Kastner P.
        • Grondona J.M.
        • Mark M.
        • et al.
        Genetic analysis of RXR alpha developmental function: convergence of RXR and RAR signaling pathways in heart and eye morphogenesis.
        Cell. 1994; 78: 987-1003
        • Kulesz-Martin M.
        • Lagowski J.
        • Fei S.
        • et al.
        Melanocyte and keratinocyte carcinogenesis: p53 family protein activities and intersecting mRNA expression profiles.
        J Investig Dermatol Symp Proc. 2005; 10: 142-152
        • Kunisada M.
        • Sakumi K.
        • Tominaga Y.
        • et al.
        8-Oxoguanine formation induced by chronic UVB exposure makes Ogg1 knockout mice susceptible to skin carcinogenesis.
        Cancer Res. 2005; 65: 6006-6010
        • Li M.
        • Chiba H.
        • Warot X.
        • et al.
        RXR-alpha ablation in skin keratinocytes results in alopecia and epidermal alterations.
        Development. 2001; 128: 675-688
        • Li M.
        • Indra A.K.
        • Warot X.
        • et al.
        Skin abnormalities generated by temporally controlled RXRalpha mutations in mouse epidermis.
        Nature. 2000; 407: 633-636
        • Lippens S.
        • Hoste E.
        • Vandenabeele P.
        • et al.
        Cell death in the skin.
        Apoptosis. 2009; 14: 549-569
        • Lu Y.P.
        • Lou Y.R.
        • Yen P.
        • et al.
        Time course for early adaptive responses to ultraviolet B light in the epidermis of SKH-1 mice.
        Cancer Res. 1999; 59: 4591-4602
        • Matsumura Y.
        • Moodycliffe A.M.
        • Nghiem D.X.
        • et al.
        Resistance of CD1d-/- mice to ultraviolet-induced skin cancer is associated with increased apoptosis.
        Am J Pathol. 2004; 165: 879-887
        • Metzger D.
        • Indra A.K.
        • Li M.
        • et al.
        Targeted conditional somatic mutagenesis in the mouse: temporally-controlled knock out of retinoid receptors in epidermal keratinocytes.
        Methods Enzymol. 2003; 364: 379-408
        • Mitchell D.L.
        • Nairn R.S.
        The biology of the (6-4) photoproduct.
        Photochem Photobiol. 1989; 49: 805-819
        • Murase D.
        • Hachiya A.
        • Amano Y.
        • et al.
        The essential role of p53 in hyperpigmentation of the skin via regulation of paracrine melanogenic cytokine receptor signaling.
        J Biol Chem. 2009; 284: 4343-4353
        • Noonan F.P.
        • Recio J.A.
        • Takayama H.
        • et al.
        Neonatal sunburn and melanoma in mice.
        Nature. 2001; 413: 271-272
        • Olivier M.
        • Hussain S.P.
        • Caron de Fromentel C.
        • et al.
        TP53 mutation spectra and load: a tool for generating hypotheses on the etiology of cancer.
        IARC Sci Publ. 2004; 157: 247-270
        • Ouhtit A.
        • Muller H.K.
        • Davis D.W.
        • et al.
        Temporal events in skin injury and the early adaptive responses in ultraviolet-irradiated mouse skin.
        Am J Pathol. 2000; 156: 201-207
        • Proksch E.
        • Brandner J.M.
        • Jensen J.M.
        The skin: an indispensable barrier.
        Exp Dermatol. 2008; 17: 1063-1072
        • Quevedo Jr, W.C.
        • Fleischmann R.D.
        Developmental biology of mammalian melanocytes.
        J Invest Dermatol. 1980; 75: 116-120
        • Rundhaug J.E.
        • Mikulec C.
        • Pavone A.
        • et al.
        A role for cyclooxygenase-2 in ultraviolet light-induced skin carcinogenesis.
        Mol Carcinog. 2007; 46: 692-698
        • Saraiya M.
        • Glanz K.
        • Briss P.A.
        • et al.
        Interventions to prevent skin cancer by reducing exposure to ultraviolet radiation: a systematic review.
        Am J Prev Med. 2004; 27: 422-466
        • Sato T.
        • Kawada A.
        Mitotic activity of hairless mouse epidermal melanocytes: its role in the increase of melanocytes during ultraviolet radiation.
        J Invest Dermatol. 1972; 58: 392-395
        • Setlow R.B.
        • Swenson P.A.
        • Carrier W.L.
        Thymine dimers and inhibition of DNA synthesis by ultraviolet irradiation of cells.
        Science. 1963; 142: 1464-1466
        • Slominski A.
        • Paus R.
        Melanogenesis is coupled to murine anagen: toward new concepts for the role of melanocytes and the regulation of melanogenesis in hair growth.
        J Invest Dermatol. 1993; 101: 90S-97S
        • Song X.
        • Mosby N.
        • Yang J.
        • et al.
        alpha-MSH activates immediate defense responses to UV-induced oxidative stress in human melanocytes.
        Pigment Cell Melanoma Res. 2009; 22: 809-818
        • Stierner U.
        • Rosdahl I.
        • Augustsson A.
        • et al.
        UVB irradiation induces melanocyte increase in both exposed and shielded human skin.
        J Invest Dermatol. 1989; 92: 561-564
        • Strom S.S.
        • Yamamura Y.
        Epidemiology of nonmelanoma skin cancer.
        Clin Plast Surg. 1997; 24: 627-636
        • Sucov H.M.
        • Dyson E.
        • Gumeringer C.L.
        • et al.
        RXR alpha mutant mice establish a genetic basis for vitamin A signaling in heart morphogenesis.
        Genes Dev. 1994; 8: 1007-1018
        • Tripp C.S.
        • Blomme E.A.
        • Chinn K.S.
        • et al.
        Epidermal COX-2 induction following ultraviolet irradiation: suggested mechanism for the role of COX-2 inhibition in photoprotection.
        J Invest Dermatol. 2003; 121: 853-861
        • van Schanke A.
        • Jongsma M.J.
        • Bisschop R.
        • et al.
        Single UVB overexposure stimulates melanocyte proliferation in murine skin, in contrast to fractionated or UVA-1 exposure.
        J Invest Dermatol. 2005; 124: 241-247
        • Walker G.J.
        • Kimlin M.G.
        • Hacker E.
        • et al.
        Murine neonatal melanocytes exhibit a heightened proliferative response to ultraviolet radiation and migrate to the epidermal basal layer.
        J Invest Dermatol. 2009; 129: 184-193
        • Werner S.
        • Smola H.
        Paracrine regulation of keratinocyte proliferation and differentiation.
        Trends Cell Biol. 2001; 11: 143-146
        • Whiteman D.C.
        • Whiteman C.A.
        • Green A.C.
        Childhood sun exposure as a risk factor for melanoma: a systematic review of epidemiologic studies.
        Cancer Causes Control. 2001; 12: 69-82
        • Wilgus T.A.
        • Parrett M.L.
        • Ross M.S.
        • et al.
        Inhibition of ultraviolet light B-induced cutaneous inflammation by a specific cyclooxygenase-2 inhibitor.
        Adv Exp Med Biol. 2002; 507: 85-92
        • Wolnicka-Glubisz A.
        • Noonan F.P.
        Neonatal susceptibility to UV induced cutaneous malignant melanoma in a mouse model.
        Photochem Photobiol Sci. 2006; 5: 254-260
        • Wulff B.C.
        • Schick J.S.
        • Thomas-Ahner J.M.
        • et al.
        Topical treatment with OGG1 enzyme affects UVB-induced skin carcinogenesis.
        Photochem Photobiol. 2008; 84: 317-321
        • Yada Y.
        • Higuchi K.
        • Imokawa G.
        Effects of endothelins on signal transduction and proliferation in human melanocytes.
        J Biol Chem. 1991; 266: 18352-18357
        • Yasui H.
        • Sakurai H.
        Chemiluminescent detection and imaging of reactive oxygen species in live mouse skin exposed to UVA.
        Biochem Biophys Res Commun. 2000; 269: 131-136
        • Yoshida A.
        • Kanno H.
        • Watabe D.
        • et al.
        The role of heparin-binding EGF-like growth factor and amphiregulin in the epidermal proliferation of psoriasis in cooperation with TNFalpha.
        Arch Dermatol Res. 2008; 300: 37-45
        • Young A.R.
        Cumulative effects of ultraviolet radiation on the skin: cancer and photoaging.
        Semin Dermatol. 1990; 9: 25-31