- Email: [email protected]
In vivo relevance for photoprotection by the vitamin D rapid response pathway
Journal of Steroid Biochemistry & Molecular Biology 103 (2007) 451–456
In vivo relevance for photoprotection by the vitamin D rapid response pathway K.M. Dixon a , S.S. Deo a , A.W. Norman b , J.E. Bishop b , G.M. Halliday c , V.E. Reeve d , R.S. Mason a,∗ a
Department of Physiology and Bosch Institute, University of Sydney, NSW 2006, Australia b Department of Biochemistry, University of California, Riverside, CA 92521, USA c Department of Medicine (Dermatology), University of Sydney, NSW 2006, Australia d Department of Veterinary Science, University of Sydney, NSW 2006, Australia
Abstract Vitamin D is produced by exposure of 7-dehydrocholesterol in the skin to UV irradiation (UVR) and further converted in the skin to the biologically active metabolite, 1,25-dihydroxyvitamin D3 (1,25(OH)2 D3 ) and other compounds. UVR also results in DNA damage producing cyclobutane pyrimidine dimers (CPD). We previously reported that 1,25(OH)2 D3 at picomolar concentrations, protects human skin cells from UVR-induced apoptosis, and decreases CPD in surviving cells. 1,25(OH)2 D3 has been shown to generate biological responses via two pathways—the classical steroid receptor/genomic pathway or a rapid, non-genomic pathway mediated by a putative membrane receptor. Whether the rapid response pathway is physiologically relevant is unclear. A cis-locked, rapid-acting agonist 1,25(OH)2 lumisterol3 (JN), entirely mimicked the actions of 1,25(OH)2 D3 to reduce fibroblast and keratinocyte loss and CPD damage after UVR. The effects of 1,25(OH)2 D3 were abolished by a rapid-acting antagonist, but not by a genomic antagonist. Skh:hr1 mice exposed to three times the minimal erythemal dose of solar-simulated UVR and treated topically with 1,25(OH)2 D3 or JN immediately after UVR showed reduction in UVR-induced UVR-induced sunburn cells (p < 0.01 and <0.05, respectively), CPD (p < 0.01 for both) and immunosuppression (p < 0.001 for both) compared with vehicle-treated mice. These results show for the first time an in vivo biological response mediated by a rapid-acting analog of the vitamin D system. The data support the hypothesis that 1,25(OH)2 D3 exerts its photoprotective effects via the rapid pathway and raise the possibility that other D compounds produced in skin may contribute to the photoprotective effects. © 2006 Elsevier Ltd. All rights reserved. Keywords: Vitamin D; 1,25-Dihydroxyvitamin D3 ; Vitamin D analogs; Cyclobutane pyrimidine dimers (CPD); Sunburn cells; Ultraviolet radiation; Immunosuppression; Skh:hr1 hairless mice
1. Introduction Exposure to ultraviolet radiation (UVR) from sunlight can result in a variety of responses in the skin including mutagenic DNA damage such as cyclobutane pyrimidine dimers (CPD) [1] and immunosuppression [2], both of which play a role in skin carcinogenesis. UVR is also the catalyst for the formation of vitamin D3 in skin. Upon the interaction of UVR with 7-dehydrocholesterol in skin, pre-vitamin D3 is formed, which thermally isomerizes to vitamin D3 [3]. Vitamin D3 undergoes two sequential hydroxylation reactions ∗
Corresponding author. E-mail address: [email protected] (R.S. Mason).
0960-0760/$ – see front matter © 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.jsbmb.2006.11.016
in the liver, followed by the kidney, forming the biologically active metabolite 1,25-dihydroxyvitamin D3 (1,25(OH)2 D3 ). There is evidence that the entire pathway to 1,25(OH)2 D3 is present in skin [4]. Previous studies in this laboratory have shown that 1,25(OH)2 D3 not only reduces the level of UVR-induced cell death in human skin cells, but also reduces CPD in surviving cells in a dose-dependent manner when added to skin cell cultures prior to and/or after UVR [5–7]. The various biological responses produced by 1,25(OH)2 D3 appear to be mediated by two major pathways. The well-studied genomic pathway depends on the interaction between 1,25(OH)2 D3 and the nuclear vitamin D receptor (VDR) followed by interaction of this steroid receptor
452
K.M. Dixon et al. / Journal of Steroid Biochemistry & Molecular Biology 103 (2007) 451–456
Fig. 1. Structures of 1,25(OH)2 -lumisterol3 (JN) and 1,25(OH)2 D3 in cis- and trans-form.
complex with vitamin D response elements present in the promoter regions of the target genes, resulting in up- or downregulation of transcription [8]. There is now considerable evidence to suggest that 1,25(OH)2 D3 can also utilize other signal transduction mechanisms in order to generate rapid, non-genomic responses. A rapid-acting, non-genomic pathway for 1,25(OH)2 D3 action has been proposed to generate a variety of cell specific responses within seconds to minutes. This involves the hormone binding to a membrane receptor, the identity of which is still not clear [9,10]. In the current study, vitamin D analogs were used to differentiate between the vitamin D pathways. The steroid hormone, 1,25(OH)2 D3 , because it is conformationally flexible, is able to generate biological responses either by the classical genomic pathway or by the membrane receptor-initiated rapid response path. It has been shown that analogs permanently locked in the cisconfiguration, as in 1␣,25(OH)2 lumisterol3 (JN), can activate only the rapid response pathway [11] (Fig. 1). The compound 1,25(OH)2 D3 (HL) acts only as a rapid response antagonist, whilst (23S)-25-dehydro-1␣-hydroxyvitamin D3 26,23-lactone (TEI-9647) acts as a genomic antagonist. No specific agonist of the genomic pathway has been reported. In our preliminary studies we showed that the photoprotective effects of 1,25(OH)2 D3 in human skin cells were likely to be acting via the rapid pathway [6], and showed that the photoprotective effects of 1,25(OH)2 D3 were present in vivo. We now provide further data to support these proposals, and show for the first time, an in vivo photoprotective effect of a low-calcemic analog that can only activate the rapid pathway.
2. Methods Fibroblasts and keratinocytes from human neonatal foreskins were cultured in 96-well plates on 5 mm glass coverslips coated with poly-l-lysine as described in [12]. Cells were treated with 1,25(OH)2 D3 or analogs 24 h prior to irradiation and media containing treatments were added immediately after UVR. Treatments were replaced with Martinez buffer solution containing d-glucose for the duration of irradiation.
The sham-irradiated plate was subjected to similar conditions but covered in aluminium foil for the duration of irradiation. The UVR source for in vitro studies consisted of one UVA and one UVB fluorescent lamp, with irradiance of 203 mJ/cm2 UVB and 1168 mJ/cm2 UVA (Phillips, Amsterdam, Holland), filtered through 0.5 mm cellulose tri-acetate (Eastman Chemical Products, Kingsport, TN) to eliminate wavelengths below 290 nm [6,7]. Cell loss after UVR was measured as previously described [7]. Cyclobutane pyrimidine dimers (CPD) were detected by immunohistochemistry and image analysis as described previously [7]. Female Skh:hr1 (albino, inbred) hairless mice were used for in vivo studies. Housing and irradiation conditions were as previously described [7]. Mice were subjected to three times the minimal erythemal dose (MED) of UVR (3.98 kJ/m2 UVB and 63.8 kJ/m2 UVA). Treatments were diluted in a base lotion containing ethanol, propylene glycol and water at a final solvent ratio of 2:1:1. Mice were treated with vehicle, 1,25(OH)2 D3 , or the low calcemic, rapid-acting analog JN. All treatments were applied dorsally, once, immediately after irradiation only. Skin samples were taken from UV-irradiated back skin 24 h post-UVR, fixed in Histochoice (Amresco) and paraffinembedded for sunburn cell analysis by routine haematoxylin and eosin staining. Sunburn cells were identified under a light microscope as cells with pyknotic nuclei surrounded by an eosinophilic cytoplasm, and were counted along the length of each skin section. Detection of CPD was by immunohistochemical analysis as previously described [7]. The contact hypersensitivity response was performed as previously described [7]. Briefly, mice were sensitized 1 week after irradiation with 100 L of 2% oxazolone (Sigma, St. Louis, MO, USA) (w/v) in absolute alcohol applied to nonirradiated abdominal skin. Sensitization was repeated on the subsequent day. The sensitized mice were challenged 2 weeks after irradiation. Ear thickness measurements were recorded before challenge and at 20 h after challenge. Results are based on quadruplicates of each treatment from experiments performed at least twice with similar results. Comparisons between treatment groups were made by one-
K.M. Dixon et al. / Journal of Steroid Biochemistry & Molecular Biology 103 (2007) 451–456
453
Fig. 2. Protective effects of 1,25(OH)2 D3 on UVR-induced skin cell loss (a) and CPD (b) reversed by rapid-acting antagonist HL but not by genomic antagonist TEI-9647. Cells were treated for 24 h with vehicle, 1,25(OH)2 D3 , TEI-9647, HL, or a combination of 1,25(OH)2 D3 and either TEI-9647 or HL. Treatments were replaced with Martinez solution for the duration of irradiation, and post-treatments added immediately after irradiation. Cell loss was determined 24 h after UVR (a). Significantly different from vehicle: *** p < 0.001 and * p < 0.05. CPD were detected 3 h after UVR (b). Significantly different from vehicle: ** p < 0.01.
way analysis of variance (ANOVA) using the GraphPad Instat statistical program (GraphPad Software Inc., San Diego, CA).
3. Results The level of UVR-induced cell loss was reduced in cells treated with either 1,25(OH)2 D3 or JN, compared with vehicle-treated cells. In pooled data from two separate experiments, mean cell loss in vehicle-treated skin fibroblasts was 37.1 ± 4.0%. This was significantly reduced to 15.4 ± 4.5% in cells treated with 1,25(OH)2 D3 (p < 0.01). The rapidacting, low-calcemic analog JN completely mimicked the effects of 1,25(OH)2 D3 , reducing cell loss to 15.2 ± 5.8% (p < 0.01). Similar results were obtained in keratinocytes. Furthermore, both 1,25(OH)2 D3 and JN significantly reduced CPD in keratinocytes from 18.6 ± 6.0 to 3.7 ± 1.0 (p < 0.01) and 5.3 ± 1.7 (p < 0.05), respectively.
As shown in Fig. 2a, the addition of the genomic pathway antagonist, TEI-9647 to skin fibroblasts had no effect on the photoprotective action of 1,25(OH)2 D3 . Mean cell loss in vehicle-treated skin fibroblasts was 36.3 ± 3.0%. This was reduced to 21.0 ± 2.2% in cells treated with 1,25(OH)2 D3 (p < 0.001). Mean loss in cells treated with TEI-9647 or HL alone was not significantly different to that of vehicle-treated cells. When cells were treated with a combination of TEI9647 and 1,25(OH)2 D3 , mean cell loss was 28.8 ± 1.7%, significantly lower than in vehicle-treated cells (p < 0.05) and not significantly different to mean loss in cells treated with 1,25(OH)2 D3 alone. In contrast, the protective effect of 1,25(OH)2 D3 on cell loss was abolished in the presence of HL (Fig. 2a). The proportion of positive CPD staining in cells treated with a combination of TEI-9647 and 1,25(OH)2 D3 did not differ from that in cells treated with 1,25(OH)2 D3 alone (Fig. 2b). In cells treated with the rapid response pathway
454
K.M. Dixon et al. / Journal of Steroid Biochemistry & Molecular Biology 103 (2007) 451–456
Fig. 3. Effects of rapid response pathway antagonist HL and genomic pathway antagonist TEI-9647 on UV-induced CPD in skin fibroblasts. Images are representative of fields captured during image analysis for (A) UV-irradiated vehicle; (B) UV-irradiated 1␣,25(OH)2 D3 ; (C) UV-irradiated Tei-9647; (D) UV-irradiated Tei-9647 + 1␣,25(OH)2 D3 ; (E) UV-irradiated HL; (F) UV-irradiated HL + 1␣,25(OH)2 D3 .
antagonist HL, there was no difference in the level of CPD compared with vehicle-treated cells. When cells were treated with a combination of HL and 1,25(OH)2 D3 , the inhibitory effects of 1,25(OH)2 D3 on CPD were abolished (Figs. 2b and 3). Sunburn cells (apoptotic keratinocytes) in mouse skin were reduced by both 1,25(OH)2 D3 and JN. In mice treated with 1,25(OH)2 D3 at doses of 22.8 and 4.6 pmol/cm2 , the numbers of sunburn cells per 245 linear micrometres of skin were reduced from 0.5 ± 0.1 to 0.2 ± 0.1 (p < 0.01) and 0.4 ± 0.1 (p < 0.05), respectively. In mice treated with JN at a dose of 22.8 pmol/cm2 , sunburn cells were reduced to 0.3 ± 0.0 (p < 0.05) (Fig. 4). Topical 1,25(OH)2 D3 or JN at similar doses reduced CPD measured 24 h post-UVR from 9 ± 3% to 3 ± 1% (p < 0.01) and 4 ± 2% (p < 0.01), respectively.
1,25(OH)2 D3 as well as JN, significantly reduced levels of UVR-induced immunosuppression in Skh:hr1 mice. As shown in Fig. 5, the level of immunosuppression in vehicle-treated mice at 20 h post-challenge was 23.4 ± 1.3%. In mice treated with 1,25(OH)2 D3 at a dose of 22.8 pmol/cm2 this was reduced to 3.9 ± 1.5% (p < 0.001). JN at the same dose (22.8 pmol/cm2 ) also significantly reduced immunosuppression at 20 h post-challenge to −3.0 ± 1.5% (p < 0.001) (Fig. 5).
4. Discussion We have previously provided preliminary evidence to support the proposal that the rapid pathway is involved in the photoprotective effect of 1,25(OH)2 D3 [6,7]. In the current
K.M. Dixon et al. / Journal of Steroid Biochemistry & Molecular Biology 103 (2007) 451–456
455
Fig. 4. Reduction in sunburn cells (apoptotic keratinocytes) by 1,25(OH)2 D3 in Skh:hr1 mouse skin. Mice were exposed to 3 MED solar-simulated UVR and treated topically with vehicle or 1,25(OH)2 D3 immediately after exposure. Skin biopsies were taken 24 h after UVR. Results presented as mean number of sunburn cells per 245 linear micrometres of skin section, n = 3. Significantly different from vehicle: ** p < 0.01 and * p < 0.05.
study, we show that the rapid-acting analog, JN, entirely mimics the photoprotective effects of 1,25(OH)2 D3 in human skin cells. Furthermore, both the reduction in UVR-induced cell loss and in CPD damage by 1,25(OH)2 D3 were fully reversed by the rapid response antagonist HL and unaffected by the genomic antagonist TEI-9647. We have shown for the first time an in vivo complex biological response using an analog that is only capable of acting via the rapid pathway. The cis-locked compound JN, which has no transcriptional activity, entirely mimicked the effects of 1,25(OH)2 D3 at equivalent doses by reducing UVR-induced sunburn cells, DNA damage and immunosuppression in Skh:hr1 hairless mice. These results provide strong and consistent evidence for a physiological role of the rapid response pathway in photoprotection.
Fig. 5. Reduction in UVR-induced systemic immunosuppression by 1,25(OH)2 D3 and JN in Skh:hr1 mice. Mice were exposed to 3 MED solarsimulated UVR and treated immediately after exposure once with vehicle, 1,25(OH)2 D3 or JN. Mice were sensitized on non-irradiated abdominal skin 7 and 8 d after UVR with 2% oxazolone. Mice were challenged on ears 7 d after sensitization and ear swelling recorded 20 h later. Results were calculated as the difference between pre- and post-challenge ear thickness measurements of non-irradiated mice as a proportion of the difference between pre- and post-challenge ear thickness measurements of irradiated mice for each treatment. Immunosuppression expressed as 100% minus this value, ± S.E.M. *** p < 0.001 significantly different from vehicle. n = 5.
These functions are likely to be important, as there is ample evidence that both production of CPD and immunosuppression are key elements in skin carcinogenesis [2]. Since CPD are produced by the direct interaction of UVR with DNA, and the protective effects of vitamin D compounds are seen following application after irradiation, it is reasonable to speculate that the CPD reduction is the result, at least in part, of increased repair, probably in turn a result of increased p53 expression and reduced nitric oxide products, previously demonstrated in the presence of 1,25(OH)2 D3 [7]. Repair of CPDs reduces skin cell apoptosis, and prevents UVRinduced mutation but also immunosuppression [13], thus providing a link between the three effects described in vivo. The process of vitamin D3 production in skin following exposure to UVR takes several hours. The conversion of vitamin D3 to 1,25(OH)2 D3 takes place over several further hours [14]. Because of this long time-course, it is unlikely that the 1,25(OH)2 D3 produced locally in skin protects from the UVR exposure responsible for its production. It is more likely that the locally produced vitamin D compounds protect from further UVR exposure, as is the case for UVR-induced increases in cornification and pigmentation [15,16]. Overirradiation products have some structural similarities to the rapid-acting agonists [3] and it is possible to speculate that these compounds may also contribute to the photoprotective effect. The data presented are consistent with the proposal that locally produced 1,25(OH)2 D3 contributes to endogenous photoprotection in skin. The data also provide evidence that the rapid response pathway for vitamin D plays an important physiological role in a whole animal.
Acknowledgements This work was supported by the National Health and Medical Research Council of Australia and by the Cancer Council of New South Wales. K.M. Dixon was the recipient
456
K.M. Dixon et al. / Journal of Steroid Biochemistry & Molecular Biology 103 (2007) 451–456
of a N.S.W. Cancer Institute Research Scholar Award. With thanks to Dr. S. Ishizuka, The Teijin Institute for Biomedical Research, Tokyo, Japan, for providing the TEI-9647 compound.
References [1] R.B. Setlow, Cyclobutane-type pyrimidine dimers in polynucleotides, Science 153 (734) (1966) 379–386. [2] V.E. Reeve, R.D. Ley, Animal models of ultraviolet radiation-induced skin cancer, in: D. Hill, J.M. Elwood, D.R. English (Eds.), Prevention of Skin Cancer. Cancer Prevention–Cancer Causes, vol. 3, Kluwer Acad Publ., Dordrecht, 2004, pp. 177–194. [3] M.F. Holick, McCollum Award Lecture, 1994: vitamin D—new horizons for the 21st century, Am. J. Clin. Nutr. 60 (4) (1994) 619– 630. [4] B. Lehmann, et al., UVB-induced conversion of 7-dehydrocholesterol to 1alpha,25-dihydroxyvitamin D3 in an in vitro human skin equivalent model, J. Invest. Dermatol. 117 (5) (2001) 1179–1185. [5] R.S. Mason, C.J. Holliday, R. Gupta, 1,25-Dihydroxyvitamin D and photoprotection in skin cells, in: D. Tsambos, H. Merk (Eds.), Modern Trends in Skin Pharmacology, Parissianos Medical Publications S.A. Athens, Athens, Greece, 2002, pp. 59–66. [6] G. Wong, et al., 1,25-Dihydroxyvitamin D and three low-calcemic analogs decrease UV-induced DNA damage via the rapid response pathway, J. Steroid Biochem. Mol. Biol. 89–90 (1–5) (2004) 567– 570. [7] K.M. Dixon, et al., Skin cancer prevention: a possible role of 1,25dihydroxyvitamin D3 and its analogs, J. Steroid Biochem. Mol. Biol. 97 (1–2) (2005) 137–143.
[8] A.W. Norman, et al., Different shapes of the steroid hormone 1alpha,25(OH)(2)-vitamin D(3) act as agonists for two different receptors in the vitamin D endocrine system to mediate genomic and rapid responses, Steroids 66 (3–5) (2001) 147–158. [9] A.W. Norman, M.T. Mizwicki, D.P. Norman, Steroid-hormone rapid actions, membrane receptors and a conformational ensemble model, Nat. Rev. 3 (1) (2004) 27–41. [10] I. Nemere, et al., Ribozyme knockdown functionally links a 1,25(OH)2 D3 membrane binding protein (1,25D3 -MARRS) and phosphate uptake in intestinal cells, Proc. Natl. Acad. Sci. U.S.A. 101 (19) (2004) 7392–7397. [11] A.W. Norman, et al., Comparison of 6-s-cis- and 6-s-trans-locked analogs of 1alpha,25-dihydroxyvitamin D3 indicates that the 6-s-cis conformation is preferred for rapid nongenomic biological responses and that neither 6-s-cis- nor 6-s-trans-locked analogs are preferred for genomic biological responses, Mol. Endocrinol. 11 (10) (1997) 1518–1531. [12] C. Gordon-Thomson, R.S. Mason, G.P. Moore, Regulation of epidermal growth factor receptor expression in human melanocytes, Exp. Dermatol. 10 (5) (2001) 321–328. [13] M.L. Kripke, et al., Pyrimidine dimers in DNA initiate systemic immunosuppression in UV-irradiated mice, Proc. Natl. Acad. Sci. U.S.A. 89 (16) (1992) 7516–7520. [14] B. Lehmann, S. Abraham, M. Meurer, Role for tumor necrosis factor-alpha in UVB-induced conversion of 7-dehydrocholesterol to 1alpha,25-dihydroxyvitamin D3 in cultured keratinocytes, J. Steroid Biochem. Mol. Biol. 89–90 (1–5) (2004) 561–565. [15] T.B. Fitzpatrick, Ultraviolet-induced pigmentary changes: benefits and hazards, Curr. Problems Dermatol. 15 (1986) 25–38. [16] N.S. Dissanayake, R.S. Mason, Modulation of skin cell functions by transforming growth factor-beta1 and ACTH after ultraviolet irradiation, J. Endocrinol. 159 (1) (1998) 153–163.
In vivo relevance for photoprotection by the vitamin D rapid response pathway K.M. Dixon a , S.S. Deo a , A.W. Norman b , J.E. Bishop b , G.M. Halliday c , V.E. Reeve d , R.S. Mason a,∗ a
Department of Physiology and Bosch Institute, University of Sydney, NSW 2006, Australia b Department of Biochemistry, University of California, Riverside, CA 92521, USA c Department of Medicine (Dermatology), University of Sydney, NSW 2006, Australia d Department of Veterinary Science, University of Sydney, NSW 2006, Australia
Abstract Vitamin D is produced by exposure of 7-dehydrocholesterol in the skin to UV irradiation (UVR) and further converted in the skin to the biologically active metabolite, 1,25-dihydroxyvitamin D3 (1,25(OH)2 D3 ) and other compounds. UVR also results in DNA damage producing cyclobutane pyrimidine dimers (CPD). We previously reported that 1,25(OH)2 D3 at picomolar concentrations, protects human skin cells from UVR-induced apoptosis, and decreases CPD in surviving cells. 1,25(OH)2 D3 has been shown to generate biological responses via two pathways—the classical steroid receptor/genomic pathway or a rapid, non-genomic pathway mediated by a putative membrane receptor. Whether the rapid response pathway is physiologically relevant is unclear. A cis-locked, rapid-acting agonist 1,25(OH)2 lumisterol3 (JN), entirely mimicked the actions of 1,25(OH)2 D3 to reduce fibroblast and keratinocyte loss and CPD damage after UVR. The effects of 1,25(OH)2 D3 were abolished by a rapid-acting antagonist, but not by a genomic antagonist. Skh:hr1 mice exposed to three times the minimal erythemal dose of solar-simulated UVR and treated topically with 1,25(OH)2 D3 or JN immediately after UVR showed reduction in UVR-induced UVR-induced sunburn cells (p < 0.01 and <0.05, respectively), CPD (p < 0.01 for both) and immunosuppression (p < 0.001 for both) compared with vehicle-treated mice. These results show for the first time an in vivo biological response mediated by a rapid-acting analog of the vitamin D system. The data support the hypothesis that 1,25(OH)2 D3 exerts its photoprotective effects via the rapid pathway and raise the possibility that other D compounds produced in skin may contribute to the photoprotective effects. © 2006 Elsevier Ltd. All rights reserved. Keywords: Vitamin D; 1,25-Dihydroxyvitamin D3 ; Vitamin D analogs; Cyclobutane pyrimidine dimers (CPD); Sunburn cells; Ultraviolet radiation; Immunosuppression; Skh:hr1 hairless mice
1. Introduction Exposure to ultraviolet radiation (UVR) from sunlight can result in a variety of responses in the skin including mutagenic DNA damage such as cyclobutane pyrimidine dimers (CPD) [1] and immunosuppression [2], both of which play a role in skin carcinogenesis. UVR is also the catalyst for the formation of vitamin D3 in skin. Upon the interaction of UVR with 7-dehydrocholesterol in skin, pre-vitamin D3 is formed, which thermally isomerizes to vitamin D3 [3]. Vitamin D3 undergoes two sequential hydroxylation reactions ∗
Corresponding author. E-mail address: [email protected] (R.S. Mason).
0960-0760/$ – see front matter © 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.jsbmb.2006.11.016
in the liver, followed by the kidney, forming the biologically active metabolite 1,25-dihydroxyvitamin D3 (1,25(OH)2 D3 ). There is evidence that the entire pathway to 1,25(OH)2 D3 is present in skin [4]. Previous studies in this laboratory have shown that 1,25(OH)2 D3 not only reduces the level of UVR-induced cell death in human skin cells, but also reduces CPD in surviving cells in a dose-dependent manner when added to skin cell cultures prior to and/or after UVR [5–7]. The various biological responses produced by 1,25(OH)2 D3 appear to be mediated by two major pathways. The well-studied genomic pathway depends on the interaction between 1,25(OH)2 D3 and the nuclear vitamin D receptor (VDR) followed by interaction of this steroid receptor
452
K.M. Dixon et al. / Journal of Steroid Biochemistry & Molecular Biology 103 (2007) 451–456
Fig. 1. Structures of 1,25(OH)2 -lumisterol3 (JN) and 1,25(OH)2 D3 in cis- and trans-form.
complex with vitamin D response elements present in the promoter regions of the target genes, resulting in up- or downregulation of transcription [8]. There is now considerable evidence to suggest that 1,25(OH)2 D3 can also utilize other signal transduction mechanisms in order to generate rapid, non-genomic responses. A rapid-acting, non-genomic pathway for 1,25(OH)2 D3 action has been proposed to generate a variety of cell specific responses within seconds to minutes. This involves the hormone binding to a membrane receptor, the identity of which is still not clear [9,10]. In the current study, vitamin D analogs were used to differentiate between the vitamin D pathways. The steroid hormone, 1,25(OH)2 D3 , because it is conformationally flexible, is able to generate biological responses either by the classical genomic pathway or by the membrane receptor-initiated rapid response path. It has been shown that analogs permanently locked in the cisconfiguration, as in 1␣,25(OH)2 lumisterol3 (JN), can activate only the rapid response pathway [11] (Fig. 1). The compound 1,25(OH)2 D3 (HL) acts only as a rapid response antagonist, whilst (23S)-25-dehydro-1␣-hydroxyvitamin D3 26,23-lactone (TEI-9647) acts as a genomic antagonist. No specific agonist of the genomic pathway has been reported. In our preliminary studies we showed that the photoprotective effects of 1,25(OH)2 D3 in human skin cells were likely to be acting via the rapid pathway [6], and showed that the photoprotective effects of 1,25(OH)2 D3 were present in vivo. We now provide further data to support these proposals, and show for the first time, an in vivo photoprotective effect of a low-calcemic analog that can only activate the rapid pathway.
2. Methods Fibroblasts and keratinocytes from human neonatal foreskins were cultured in 96-well plates on 5 mm glass coverslips coated with poly-l-lysine as described in [12]. Cells were treated with 1,25(OH)2 D3 or analogs 24 h prior to irradiation and media containing treatments were added immediately after UVR. Treatments were replaced with Martinez buffer solution containing d-glucose for the duration of irradiation.
The sham-irradiated plate was subjected to similar conditions but covered in aluminium foil for the duration of irradiation. The UVR source for in vitro studies consisted of one UVA and one UVB fluorescent lamp, with irradiance of 203 mJ/cm2 UVB and 1168 mJ/cm2 UVA (Phillips, Amsterdam, Holland), filtered through 0.5 mm cellulose tri-acetate (Eastman Chemical Products, Kingsport, TN) to eliminate wavelengths below 290 nm [6,7]. Cell loss after UVR was measured as previously described [7]. Cyclobutane pyrimidine dimers (CPD) were detected by immunohistochemistry and image analysis as described previously [7]. Female Skh:hr1 (albino, inbred) hairless mice were used for in vivo studies. Housing and irradiation conditions were as previously described [7]. Mice were subjected to three times the minimal erythemal dose (MED) of UVR (3.98 kJ/m2 UVB and 63.8 kJ/m2 UVA). Treatments were diluted in a base lotion containing ethanol, propylene glycol and water at a final solvent ratio of 2:1:1. Mice were treated with vehicle, 1,25(OH)2 D3 , or the low calcemic, rapid-acting analog JN. All treatments were applied dorsally, once, immediately after irradiation only. Skin samples were taken from UV-irradiated back skin 24 h post-UVR, fixed in Histochoice (Amresco) and paraffinembedded for sunburn cell analysis by routine haematoxylin and eosin staining. Sunburn cells were identified under a light microscope as cells with pyknotic nuclei surrounded by an eosinophilic cytoplasm, and were counted along the length of each skin section. Detection of CPD was by immunohistochemical analysis as previously described [7]. The contact hypersensitivity response was performed as previously described [7]. Briefly, mice were sensitized 1 week after irradiation with 100 L of 2% oxazolone (Sigma, St. Louis, MO, USA) (w/v) in absolute alcohol applied to nonirradiated abdominal skin. Sensitization was repeated on the subsequent day. The sensitized mice were challenged 2 weeks after irradiation. Ear thickness measurements were recorded before challenge and at 20 h after challenge. Results are based on quadruplicates of each treatment from experiments performed at least twice with similar results. Comparisons between treatment groups were made by one-
K.M. Dixon et al. / Journal of Steroid Biochemistry & Molecular Biology 103 (2007) 451–456
453
Fig. 2. Protective effects of 1,25(OH)2 D3 on UVR-induced skin cell loss (a) and CPD (b) reversed by rapid-acting antagonist HL but not by genomic antagonist TEI-9647. Cells were treated for 24 h with vehicle, 1,25(OH)2 D3 , TEI-9647, HL, or a combination of 1,25(OH)2 D3 and either TEI-9647 or HL. Treatments were replaced with Martinez solution for the duration of irradiation, and post-treatments added immediately after irradiation. Cell loss was determined 24 h after UVR (a). Significantly different from vehicle: *** p < 0.001 and * p < 0.05. CPD were detected 3 h after UVR (b). Significantly different from vehicle: ** p < 0.01.
way analysis of variance (ANOVA) using the GraphPad Instat statistical program (GraphPad Software Inc., San Diego, CA).
3. Results The level of UVR-induced cell loss was reduced in cells treated with either 1,25(OH)2 D3 or JN, compared with vehicle-treated cells. In pooled data from two separate experiments, mean cell loss in vehicle-treated skin fibroblasts was 37.1 ± 4.0%. This was significantly reduced to 15.4 ± 4.5% in cells treated with 1,25(OH)2 D3 (p < 0.01). The rapidacting, low-calcemic analog JN completely mimicked the effects of 1,25(OH)2 D3 , reducing cell loss to 15.2 ± 5.8% (p < 0.01). Similar results were obtained in keratinocytes. Furthermore, both 1,25(OH)2 D3 and JN significantly reduced CPD in keratinocytes from 18.6 ± 6.0 to 3.7 ± 1.0 (p < 0.01) and 5.3 ± 1.7 (p < 0.05), respectively.
As shown in Fig. 2a, the addition of the genomic pathway antagonist, TEI-9647 to skin fibroblasts had no effect on the photoprotective action of 1,25(OH)2 D3 . Mean cell loss in vehicle-treated skin fibroblasts was 36.3 ± 3.0%. This was reduced to 21.0 ± 2.2% in cells treated with 1,25(OH)2 D3 (p < 0.001). Mean loss in cells treated with TEI-9647 or HL alone was not significantly different to that of vehicle-treated cells. When cells were treated with a combination of TEI9647 and 1,25(OH)2 D3 , mean cell loss was 28.8 ± 1.7%, significantly lower than in vehicle-treated cells (p < 0.05) and not significantly different to mean loss in cells treated with 1,25(OH)2 D3 alone. In contrast, the protective effect of 1,25(OH)2 D3 on cell loss was abolished in the presence of HL (Fig. 2a). The proportion of positive CPD staining in cells treated with a combination of TEI-9647 and 1,25(OH)2 D3 did not differ from that in cells treated with 1,25(OH)2 D3 alone (Fig. 2b). In cells treated with the rapid response pathway
454
K.M. Dixon et al. / Journal of Steroid Biochemistry & Molecular Biology 103 (2007) 451–456
Fig. 3. Effects of rapid response pathway antagonist HL and genomic pathway antagonist TEI-9647 on UV-induced CPD in skin fibroblasts. Images are representative of fields captured during image analysis for (A) UV-irradiated vehicle; (B) UV-irradiated 1␣,25(OH)2 D3 ; (C) UV-irradiated Tei-9647; (D) UV-irradiated Tei-9647 + 1␣,25(OH)2 D3 ; (E) UV-irradiated HL; (F) UV-irradiated HL + 1␣,25(OH)2 D3 .
antagonist HL, there was no difference in the level of CPD compared with vehicle-treated cells. When cells were treated with a combination of HL and 1,25(OH)2 D3 , the inhibitory effects of 1,25(OH)2 D3 on CPD were abolished (Figs. 2b and 3). Sunburn cells (apoptotic keratinocytes) in mouse skin were reduced by both 1,25(OH)2 D3 and JN. In mice treated with 1,25(OH)2 D3 at doses of 22.8 and 4.6 pmol/cm2 , the numbers of sunburn cells per 245 linear micrometres of skin were reduced from 0.5 ± 0.1 to 0.2 ± 0.1 (p < 0.01) and 0.4 ± 0.1 (p < 0.05), respectively. In mice treated with JN at a dose of 22.8 pmol/cm2 , sunburn cells were reduced to 0.3 ± 0.0 (p < 0.05) (Fig. 4). Topical 1,25(OH)2 D3 or JN at similar doses reduced CPD measured 24 h post-UVR from 9 ± 3% to 3 ± 1% (p < 0.01) and 4 ± 2% (p < 0.01), respectively.
1,25(OH)2 D3 as well as JN, significantly reduced levels of UVR-induced immunosuppression in Skh:hr1 mice. As shown in Fig. 5, the level of immunosuppression in vehicle-treated mice at 20 h post-challenge was 23.4 ± 1.3%. In mice treated with 1,25(OH)2 D3 at a dose of 22.8 pmol/cm2 this was reduced to 3.9 ± 1.5% (p < 0.001). JN at the same dose (22.8 pmol/cm2 ) also significantly reduced immunosuppression at 20 h post-challenge to −3.0 ± 1.5% (p < 0.001) (Fig. 5).
4. Discussion We have previously provided preliminary evidence to support the proposal that the rapid pathway is involved in the photoprotective effect of 1,25(OH)2 D3 [6,7]. In the current
K.M. Dixon et al. / Journal of Steroid Biochemistry & Molecular Biology 103 (2007) 451–456
455
Fig. 4. Reduction in sunburn cells (apoptotic keratinocytes) by 1,25(OH)2 D3 in Skh:hr1 mouse skin. Mice were exposed to 3 MED solar-simulated UVR and treated topically with vehicle or 1,25(OH)2 D3 immediately after exposure. Skin biopsies were taken 24 h after UVR. Results presented as mean number of sunburn cells per 245 linear micrometres of skin section, n = 3. Significantly different from vehicle: ** p < 0.01 and * p < 0.05.
study, we show that the rapid-acting analog, JN, entirely mimics the photoprotective effects of 1,25(OH)2 D3 in human skin cells. Furthermore, both the reduction in UVR-induced cell loss and in CPD damage by 1,25(OH)2 D3 were fully reversed by the rapid response antagonist HL and unaffected by the genomic antagonist TEI-9647. We have shown for the first time an in vivo complex biological response using an analog that is only capable of acting via the rapid pathway. The cis-locked compound JN, which has no transcriptional activity, entirely mimicked the effects of 1,25(OH)2 D3 at equivalent doses by reducing UVR-induced sunburn cells, DNA damage and immunosuppression in Skh:hr1 hairless mice. These results provide strong and consistent evidence for a physiological role of the rapid response pathway in photoprotection.
Fig. 5. Reduction in UVR-induced systemic immunosuppression by 1,25(OH)2 D3 and JN in Skh:hr1 mice. Mice were exposed to 3 MED solarsimulated UVR and treated immediately after exposure once with vehicle, 1,25(OH)2 D3 or JN. Mice were sensitized on non-irradiated abdominal skin 7 and 8 d after UVR with 2% oxazolone. Mice were challenged on ears 7 d after sensitization and ear swelling recorded 20 h later. Results were calculated as the difference between pre- and post-challenge ear thickness measurements of non-irradiated mice as a proportion of the difference between pre- and post-challenge ear thickness measurements of irradiated mice for each treatment. Immunosuppression expressed as 100% minus this value, ± S.E.M. *** p < 0.001 significantly different from vehicle. n = 5.
These functions are likely to be important, as there is ample evidence that both production of CPD and immunosuppression are key elements in skin carcinogenesis [2]. Since CPD are produced by the direct interaction of UVR with DNA, and the protective effects of vitamin D compounds are seen following application after irradiation, it is reasonable to speculate that the CPD reduction is the result, at least in part, of increased repair, probably in turn a result of increased p53 expression and reduced nitric oxide products, previously demonstrated in the presence of 1,25(OH)2 D3 [7]. Repair of CPDs reduces skin cell apoptosis, and prevents UVRinduced mutation but also immunosuppression [13], thus providing a link between the three effects described in vivo. The process of vitamin D3 production in skin following exposure to UVR takes several hours. The conversion of vitamin D3 to 1,25(OH)2 D3 takes place over several further hours [14]. Because of this long time-course, it is unlikely that the 1,25(OH)2 D3 produced locally in skin protects from the UVR exposure responsible for its production. It is more likely that the locally produced vitamin D compounds protect from further UVR exposure, as is the case for UVR-induced increases in cornification and pigmentation [15,16]. Overirradiation products have some structural similarities to the rapid-acting agonists [3] and it is possible to speculate that these compounds may also contribute to the photoprotective effect. The data presented are consistent with the proposal that locally produced 1,25(OH)2 D3 contributes to endogenous photoprotection in skin. The data also provide evidence that the rapid response pathway for vitamin D plays an important physiological role in a whole animal.
Acknowledgements This work was supported by the National Health and Medical Research Council of Australia and by the Cancer Council of New South Wales. K.M. Dixon was the recipient
456
K.M. Dixon et al. / Journal of Steroid Biochemistry & Molecular Biology 103 (2007) 451–456
of a N.S.W. Cancer Institute Research Scholar Award. With thanks to Dr. S. Ishizuka, The Teijin Institute for Biomedical Research, Tokyo, Japan, for providing the TEI-9647 compound.
References [1] R.B. Setlow, Cyclobutane-type pyrimidine dimers in polynucleotides, Science 153 (734) (1966) 379–386. [2] V.E. Reeve, R.D. Ley, Animal models of ultraviolet radiation-induced skin cancer, in: D. Hill, J.M. Elwood, D.R. English (Eds.), Prevention of Skin Cancer. Cancer Prevention–Cancer Causes, vol. 3, Kluwer Acad Publ., Dordrecht, 2004, pp. 177–194. [3] M.F. Holick, McCollum Award Lecture, 1994: vitamin D—new horizons for the 21st century, Am. J. Clin. Nutr. 60 (4) (1994) 619– 630. [4] B. Lehmann, et al., UVB-induced conversion of 7-dehydrocholesterol to 1alpha,25-dihydroxyvitamin D3 in an in vitro human skin equivalent model, J. Invest. Dermatol. 117 (5) (2001) 1179–1185. [5] R.S. Mason, C.J. Holliday, R. Gupta, 1,25-Dihydroxyvitamin D and photoprotection in skin cells, in: D. Tsambos, H. Merk (Eds.), Modern Trends in Skin Pharmacology, Parissianos Medical Publications S.A. Athens, Athens, Greece, 2002, pp. 59–66. [6] G. Wong, et al., 1,25-Dihydroxyvitamin D and three low-calcemic analogs decrease UV-induced DNA damage via the rapid response pathway, J. Steroid Biochem. Mol. Biol. 89–90 (1–5) (2004) 567– 570. [7] K.M. Dixon, et al., Skin cancer prevention: a possible role of 1,25dihydroxyvitamin D3 and its analogs, J. Steroid Biochem. Mol. Biol. 97 (1–2) (2005) 137–143.
[8] A.W. Norman, et al., Different shapes of the steroid hormone 1alpha,25(OH)(2)-vitamin D(3) act as agonists for two different receptors in the vitamin D endocrine system to mediate genomic and rapid responses, Steroids 66 (3–5) (2001) 147–158. [9] A.W. Norman, M.T. Mizwicki, D.P. Norman, Steroid-hormone rapid actions, membrane receptors and a conformational ensemble model, Nat. Rev. 3 (1) (2004) 27–41. [10] I. Nemere, et al., Ribozyme knockdown functionally links a 1,25(OH)2 D3 membrane binding protein (1,25D3 -MARRS) and phosphate uptake in intestinal cells, Proc. Natl. Acad. Sci. U.S.A. 101 (19) (2004) 7392–7397. [11] A.W. Norman, et al., Comparison of 6-s-cis- and 6-s-trans-locked analogs of 1alpha,25-dihydroxyvitamin D3 indicates that the 6-s-cis conformation is preferred for rapid nongenomic biological responses and that neither 6-s-cis- nor 6-s-trans-locked analogs are preferred for genomic biological responses, Mol. Endocrinol. 11 (10) (1997) 1518–1531. [12] C. Gordon-Thomson, R.S. Mason, G.P. Moore, Regulation of epidermal growth factor receptor expression in human melanocytes, Exp. Dermatol. 10 (5) (2001) 321–328. [13] M.L. Kripke, et al., Pyrimidine dimers in DNA initiate systemic immunosuppression in UV-irradiated mice, Proc. Natl. Acad. Sci. U.S.A. 89 (16) (1992) 7516–7520. [14] B. Lehmann, S. Abraham, M. Meurer, Role for tumor necrosis factor-alpha in UVB-induced conversion of 7-dehydrocholesterol to 1alpha,25-dihydroxyvitamin D3 in cultured keratinocytes, J. Steroid Biochem. Mol. Biol. 89–90 (1–5) (2004) 561–565. [15] T.B. Fitzpatrick, Ultraviolet-induced pigmentary changes: benefits and hazards, Curr. Problems Dermatol. 15 (1986) 25–38. [16] N.S. Dissanayake, R.S. Mason, Modulation of skin cell functions by transforming growth factor-beta1 and ACTH after ultraviolet irradiation, J. Endocrinol. 159 (1) (1998) 153–163.