Melanocyte-Stimulating Hormone Directly Enhances UV-Induced DNA Repair in Keratinocytes by a Xeroderma Pigmentosum Group A–Dependent Mechanism

UVB irradiation is one of the major environmental hazards in public health. Overexposure to UV leads to skin damage in short term and might result in various skin diseases and even cancers over time (1, 2). Given the difficulty in curing melanomas and other skin cancers, tremendous amounts of effort have been invested in improving the prevention of these malignancies. Studying α-melanocyte-stimulating hormone (α-MSH) signaling and learning its role in melanogenesis has suggested a potential strategy for preventing skin cancers through enhancing pigmentation with α-MSH analogues or cyclic AMP (cAMP) agonists (3–5). To date, several studies have indicated that α-MSH might reduce UV-induced skin damage via alternative mechanisms (1, 3, 6–8). Specifically, some results suggest that α-MSH directly triggers UV protective signaling in keratinocytes (9). Because keratinocytes are incapable of melanin synthesis, we wondered if α-MSH–mediated UV protection works by augmenting the DNA repair pathway.

Molecular and genetic data suggest that the melanocortin-1 receptor (MC1R) plays a crucial role in tanning and pigmentation in humans and mice (4). MC1R is expressed in melanocytes and is activated by its ligand α-MSH, a propigmentation hormone produced and secreted by both keratinocytes and melanocytes after UV irradiation (4, 10). Upon α-MSH binding, MC1R activates the cAMP pathway and melanin production in melanocytes. The gene encoding α-MSH is proopiomelanocortin (POMC), a multicomponent precursor for α-MSH (melanotropic), ACTH (adrenocorticotropic), and the opioid peptide β-endorphin (11). Mutations in the POMC gene result in a red-hair, fair-skin phenotype (like that of MC1R variant alleles) in addition to metabolic abnormalities, such as adrenal insufficiency and obesity (12).

We have shown that p53 activation in keratinocytes represents a “UV sensor/effector” for skin pigmentation, with its key mechanistic role being the transcriptional activation of POMC (9). The tumor suppressor protein p53 plays a central role in the cellular response to UV-induced DNA damage/repair (13). On the other hand, the Abdel-Malek group has shown that α-MSH can prevent UV-induced DNA damage and apoptosis in melanocytes (1, 3, 6–8). Collectively, these two observations led us to examine the possibility that α-MSH prevents UV-induced DNA damage in keratinocytes.

UV irradiation can induce DNA photoproducts. Cyclobutane pyrimidine dimers (CPD) and pyrimidine (6-4) pyrimidone photoproducts (6-4 PP) have been completely isolated and characterized as UV-induced photodimers (14–16). Among the various UV-induced DNA repair mechanisms, nucleotide excision repair (NER) represents the most versatile and flexible DNA repair pathway in cells (17). NER recognizes bulky, helix-distorting changes, such as CPD and 6-4 PPs, as well as single-strand breaks. Xeroderma pigmentosum group A (XPA) gene plays an indispensable role in the NER pathway (18). XPA was cloned from patients with xeroderma pigmentosum (XP) complementation group A (19). XP is an autosomal recessive human disease that causes sensitivity to sunlight. Patients with XP have a high incidence of skin cancer on sun-exposed skin caused by loss of the ability to repair UV-induced DNA damage, especially in the early steps of NER (20). After UVB irradiation, XPA undergoes a dramatic cytoplasm-to-nucleus translocation via association with cytoplasmic GTPase XPA-binding protein 1 (XAB1; refs. 21, 22). Once inside the nucleus, XPA is a “docking” protein for components of the NER system. XPA is involved in both global and transcription-coupled DNA repair (18, 23, 24).

Here, we showed that the level of DNA damage was remarkably reduced in epidermis that had received α-MSH before UVB irradiation in vivo. Our in vitro studies showed that, independent of its effects on melanogenesis, α-MSH protects keratinocytes from UVB-induced DNA damage. We found that the protective effect of α-MSH for UV-induced DNA damage is mediated through both XAB1 and XPA proteins. Moreover, our results showed that α-MSH treatment activates XAB1 GTPase activity, the MC1R-cAMP-Epac signaling cascade, and subsequently augments the nuclear translocation of XPA protein to promote DNA repair. Together, these studies revealed an alternative mechanism through which α-MSH protects the skin from UVB irradiation. Besides expanding our knowledge on α-MSH signaling, our results suggested potential intervention strategies to reduce UVB-induced skin damage, which might even help prevent the development of various UVB-related skin diseases.

Discarded human foreskins were collected from the Department of Obstetrics and Gynecology, Loyola University Health System. Primary keratinocytes were isolated from discarded foreskins as described previously (25) and were cultured in keratinocyte serum-free medium (Invitrogen Corporation, USA). H89 was purchased from Promega. Brefeldin A (BFA) was purchased from Sigma-Aldrich.

Whole skin samples were collected from the back of each mouse after euthanasia. A distinct separation of the epidermis from the dermis was achieved by heating the samples at 60°C for 30 seconds then placing them in ice-cold water for 1 minute and scraping off the epidermis with a single-edged razor blade (26).

Human primary keratinocytes (HPK) were deprived of bovine pituitary extract for 3 days before α-MSH treatment. Cells were preincubated with α-MSH (Phoenix, Inc.) at 1 μmol for 3 hours before UV irradiation. Cells were washed by PBS twice before UVB irradiation.

For the in vitro UV experiments, cells were exposed to UV radiation in the Stratalinker UV chamber (Stratagene) with UVB bulbs (UVP, Inc.). UV emittance was measured with the use of a UV photometer (UV Products; ref. 9). Adherent cells were irradiated through a small volume of PBS at a dose of 100 J/m².

A UVB dose of 100 J/m² is equivalent to one standard erythema dose of UVB (SED), commonly used as a measurement of sunlight. As a reference, the ambient exposure over an entire sunny summer day in Europe (Switzerland) is ∼30 to 40 SED (27).

Foreskins were injected intracutaneously with α-MSH (100 ng) for 3 hours followed by UVB irradiation (500 J/m²). After irradiation, foreskins were incubated in DMEM in humidified incubators until the time of assay. These foreskin studies have undergone Institutional Review Board (IRB) review by the IRB at Loyola University Chicago,

Sections (5 μm) of paraffin-embedded tissues were analyzed by immunofluorescence according to standard protocols with the following antibodies: anti-CPD antibody (clone MC-062) and anti–6-4 PPs antibody (clone KTM50; both from Kamiya Biomedical Company).

The immunodot blot assay used antibodies specific for CPDs and 6-4 PPs. Briefly, heat-denatured genomic DNA was dot-blotted onto a nitrocellulose membrane. After multiple washes with PBS-T, the membrane was incubated with anti-CPD antibody or anti–6-4 PPs antibody (diluted 1:1,000 in PBS-T plus 1% NFM) for 2 hours at room temperature.

Western blotting and immunoprecipitation were performed using the following antibodies: anti-XPA (Santa Cruz, Inc.), anti-XAB1 (Abcam, Inc), anti-ERCC1 (Santa Cruz, Inc.), and anti-MC1R (Alpha Diagnostic International).

Enzyme immunoassay was performed using the anti-CPD and anti–6-4 PPs antibodies as described above. Briefly, the purified antimouse IgG monoclonal antibody (mAb) was diluted to 2 μg/mL in PBS and added into an enhanced protein-binding enzyme-linked immunosorbent assay (ELISA) plate (e.g., Falcon Labware plate). MC-062 and KTM-50 mAbs specific for CPDs and 6-4 PPs were diluted 1:1,000 in blocking buffer and added to each well. Absorbance at 405 nm was determined.

Short hairpin RNA duplexes that target MC1R, XAB1, and XPA were purchased from Applied Biosystem and Santa Cruz.

All pull-down experiments were performed using Thermo (Pierce) Scientific Active GTPase Pull-Down and Detection kits with glutathione S-transferase (GST)–XAB1/XPA protein-binding domain peptide. The respective binding domain of the XPA for XAB1 GTPase was expressed as a GST-fusion protein, which, when immobilized on a resin, was used to pull down the active (i.e., GTP-bound) XAB1 GTPase. The bound GTPase was then recovered by eluting the GST-fusion protein from the glutathione resin. The purified XAB1 GTPase was finally detected by Western blot using a specific anti-XPA antibody.

Specifically, cell lysates (500 mg) were incubated with 50 mL glutathione resin and GST-XAB1 binding domain from XPA for 1 hour to capture active XAB1 GTPases. Bound XAB1 was eluted with sample buffer provided by Pierce, detected by SDS-PAGE (4–20% polyacrylamide gel), and transferred to polyvinylidene difluoride membrane for Western blotting with a specific anti-XAB1 antibody (Abcam, Inc.).

It is known that α-MSH stimulates pigment production in melanocytes and eumelanin provides a natural shield to protect skin from UV-induced damage. However, it remains unclear whether α-MSH could act directly on keratinocytes independent of its activity on pigment induction. To answer this question, we first assessed whether α-MSH could reduce UV-induced DNA damage in vivo.

UVB irradiation mainly induces two types of DNA lesions (damage), 6-4 PPs and CPD (23, 28). To evaluate the role of α-MSH in epidermis in vivo, normal human foreskin specimens were exposed to 500 J/m² UVB (9, 29) and stained for CPD and 6-4 PPs at different time points. UVB (500 J/m²) is equal to the minimal erythema dose of UVB on type IV and Indian skin, which is usually used to evaluate the UV response of human skin (29). One group received α-MSH intracutaneously 3 hours before UVB irradiation. Epidermis was collected at 3 hours after UVB irradiation to detect DNA damage by ELISA, immunofluorescence, and immunodot blot with a CPD-specific mAb and a 6-4 PPs–specific mAb (Fig. 1). CPD and 6-4 PPs were evaluated at 3 hours after UVB irradiation, because CPD became undetectable 8 hours after UVB irradiation in both groups (data not shown). As shown in Fig. 1, the number of DNA adducts was dramatically reduced in human foreskin that had received α-MSH before UVB irradiation. The same protective effect was also observed in C57BL/6 mice (Supplementary Fig. S1). Our histologic staining showed that sunburn cells were dramatically reduced in mouse epidermis that had received α-MSH before UVB irradiation (Supplementary Fig. S2). Together, these results showed that α-MSH can effectively reduce UV-induced DNA damage in both human and murine epidermis.

It is plausible that the α-MSH–mediated protection we saw in the in vivo systems above are entirely dependent on pigment induction in the melanocytes. However, it is well known that mouse melanocytes normally reside in hair follicles and dermis rather than epidermis (30). Therefore, it is highly probable that α-MSH acts directly on keratinocytes to reduce UVB-induced DNA damage. To test this hypothesis, we measured DNA damage in isolated HPKs upon UVB irradiation (100 J/m²) with or without preincubation with α-MSH (10⁻⁶ mol/L). ELISA, immunodot blot, and immunofluorescence analysis revealed that UVB-induced DNA damage was significantly reduced. Specifically, we observed >50% reduction of CPD and 6-4 PPs in the presence of α-MSH treatment (Fig. 2). Together, these results showed that α-MSH can act directly on keratinocytes to protect them from UVB-induced DNA damage.

Because keratinocytes are incapable of synthesizing melanin, we wondered how α-MSH could reduce UV-induced DNA damage in these cells. One possibility is that α-MSH might augment the DNA repair machinery in keratinocytes. Hence, we tested whether any of the DNA repair proteins are essential for α-MSH–mediated UVB protection. Specifically, we focused on XPA, a critical component of the NER pathway.

First, we explored the need for XPA protein in α-MSH protection after UV irradiation using the RNAi knockdown approach. In particular, we stably reduced the expression of XPA by specific siRNA in HPKs (Fig. 3A), and the protective effect of α-MSH was measured three hours after UVB irradiation. As shown in Fig. 3B and C, α-MSH treatment reduced the levels of CPD and 6-4 PPs in control siRNA-expressing cells. In contrast, similar levels of photoproducts CPD and 6-4 PPs were induced and cleared at the same speed with α-MSH treatment in siXPA HPK compared with untreated control cells. In other words, the protective effect of α-MSH was abolished following siXPA. Together, these results indicate that XPA is required for the protective effect of α-MSH against UV-induced DNA damage.

We then asked how α-MSH affects XPA activity to enhance DNA repair upon UVB radiation. Specifically, we assessed the affect of α-MSH on the nuclear translocation of XPA protein and the formation of nuclear foci after UVB irradiation. As shown in Fig. 4A, the total amounts of XPA in the cells remained essentially the same after UVB irradiation (100 J/m²) and α-MSH treatment (10⁻⁶ mol/L). In contrast, we observed an accumulation of XPA nuclear protein in keratinocytes, especially with UV radiation. Moreover, an increasing number of XPA nuclear foci were seen at the same time points with α-MSH treatment (Fig. 4B). We also asked whether α-MSH could enhance the binding of other repair factors to XPA to augment DNA damage repair. Specifically, we assessed the interactions between XPA and ERCC1 upon UV radiation and α-MSH treatment with immunoprecipitation followed by Western analysis. Indeed, we observed an increase in XPA and ERCC1 association upon UV radiation. The pretreatment with α-MSH further enhanced the interaction between these two proteins (Fig. 4C). Together, these results suggest that α-MSH promotes nuclear translocation of XPA and enhances its interactions with other DNA repair proteins, such as ERCC1.

It is known that XAB1 is able to bind the NH₂ terminal region of the XPA protein in cytoplasm and induces the nuclear translocation of XPA protein (21). We wondered if α-MSH–induced XPA translocation is mediated by XAB1. To test this notion, we first determined whether XAB1 is required for α-MSH protection and XPA protein translocation after UV irradiation using the siRNA knockdown approach. Specifically, we stably reduced the expression of XAB1 by siRNA in HPKs (Fig. 5A). The protective effect of α-MSH and the expression of XPA protein were measured 3 hours after UVB irradiation. As shown in Fig. 5B and C, α-MSH failed to induce DNA protection or augment nuclear translocation of XPA protein in HPKs with stable XAB1 knockdown, indicating that XAB1 is required for the protective effect of α-MSH against UV-induced DNA damage.

We then investigated whether α-MSH activates XAB1 GTPase activity using GTPase pull-down with GST-XAB1/XPA binding domain peptide followed by Western detection for activated XAB1. As shown in Fig. 5D, incubation of HPKs with α-MSH for 3 hours indeed activated XAB1 GTPase activity. These results suggest that α-MSH induces XPA nuclear translocation through activation of XAB1 GTPase activity to protect keratinocytes from UVB irradiation.

Thus far, we have shown that α-MSH reduces DNA damage in keratinocytes through activation of XAB1, which subsequently promotes XPA nuclear translocation. This mechanism seems to be completely independent of α-MSH's role in pigmentary induction in melanocytes. We then asked whether α-MSH augments DNA repair via the canonical MC1R-cAMP pathway or a different signaling cascade. First, we investigated whether α-MSH protects UV-induced damage through MC1R. As shown in Fig. 6A and Supplementary Fig. S3A and B, MC1R mRNA and protein are expressed in both HPKs and PAM212 mouse keratinocytes. The expression levels are similar to what we have observed in human primary melanocytes (HPM) or B16 melanoma cells. Next, we prepared primary keratinocytes with stable reduction of MC1R expression by specific siRNA knockdown and examined the protective effect of α-MSH 3 hours after UVB irradiation. We found that the protective effect of α-MSH was abolished following siMC1R, suggesting that MC1R is required for α-MSH protection against UV-induced DNA damage (Fig. 6B; Supplementary Fig. S3C).

Next, we explored whether cAMP signaling connects MC1R activation to the augmentation of XAB1 activity. In particular, we examined the effects of C-ANF_4-23, an adenylate cyclase inhibitor, and H89, a cAMP-dependent protein kinase (PKA) inhibitor, on the intervention of α-MSH protection against UV-induced DNA damage. As shown in Fig. 6C and Supplementary Fig. S4A, inhibition of adenylate cyclase by C-ANF_4-23 inhibits the ability of α-MSH to accelerate the clearance of UV-induced photoproducts (CPD and 6-4 PPs). α-MSH–induced/augmented XAB1GTPase activity and nuclear translocation of XPA protein were also undetectable upon inhibition of adenylate cyclase by C-ANF_4-23 (Supplementary Fig. S4B and C). Consistent with the unchanged photoproduct induction, α-MSH–induced XAB1 GTPase activity and nuclear translocation of XPA protein became undetectable in the presence of C-ANF_4-23 (Supplementary Fig. S4B and C). In contrast, α-MSH is still able to accelerate the clearance of UV-induced photoproducts (CPD and 6-4 PPs) when we inhibit PKA by H89 (Supplementary Fig. S5). Together, these results suggest that α-MSH–induced DNA repair occurs through the MC1R-cAMP signaling pathway in a PKA-independent manner. Therefore, we asked which alternative cAMP receptor could be potentially involved in this process. Specifically, we examined whether cAMP signals through Epac (exchange protein directly activated by cAMP) to activate the DNA repair machinery using an Epac antagonist, BFA (31).

As shown in Fig. 6D and Supplementary Fig. S6A, α-MSH fails to reduce UV-induced photoproducts (CPD and 6-4 PPs) if we inhibit Epac by BFA. Consistently, α-MSH–induced XAB1GTPase activity and nuclear translocation of XPA protein became undetectable in the presence of BFA (Supplementary Fig. S6B and C). These results further support the notion that α-MSH induces DNA repair through cAMP in keratinocytes and suggest Epac as the downstream effector of cAMP in this process.

α-MSH is a propigmentation hormone, which is produced and secreted by both keratinocytes and melanocytes in the skin following UV irradiation (10, 32, 33). Normal synthesis of α-MSH and ACTH is crucial for constitutive human pigmentation and the cutaneous response to UV irradiation (10, 34, 35). In addition to being the key factor in UV-induced melanogenesis, α-MSH also has crucial antiinflammatory effects in response to UV irradiation (36, 37). These data indicate that POMC/α-MSH functions as a “protective factor” against the harmful effects of UV irradiation to maintain epidermal homeostasis and genomic stability. We have further shown that the downstream target of p53 in keratinocytes (POMC/α-MSH) is also an activator of DNA repair in these cells after UV irradiation, independent of its previously known role in inducing pigmentation.

It has been previously shown by Abdel-Malek's and Schwarz's groups that α-MSH protects HPMs from UV-induced apoptosis and DNA damage (photoproducts production and oxidative stress; refs. 1, 6–8). Specifically, the antiapoptotic effects of α-MSH were mediated by the activation of the inositol triphosphate kinase–Akt pathway and microphthalmia-associated transcription factor (8). The Abdel-Malek group has also developed melanocortin analogues, which, in addition to stimulating pigmentation, effectively reduced and repaired DNA damage resulting from exposure to solar UV radiation (1, 3). Together, these studies suggest the existence of alternative mechanisms for pigmentation induction through which α-MSH protects skin from UV irradiation.

Interestingly, Abdel-Malek, Schwarz, and their colleagues have found that α-MSH does not induce the cell cycle arrest nor the activation of apoptosis-related proteins (6–8). One recent report showed that α-MSH prevented the UVB-induced suppression of a pathway functioning in protection against oxidative stress (38). In particular, it showed that UVB failed to inhibit Nrf2, a transcription factor critical to antioxidant induction, and Nrf-dependent genes upon α-MSH treatment in both keratinocytes and melanocytes (38). Complementary to these results, we showed that α-MSH is involved in the protection against UV-induced DNA damage through the activation of DNA repair machinery, including XAB1 and XPA DNA repair factors.

Several independent groups have characterized the expression of MC1R in keratinocytes (39, 40). However, the role of MC1R in keratinocytes is still unclear. Here, we showed that MC1R is required in the α-MSH–mediated UV-induced DNA repair in keratinocytes. Usually, following the binding of hormones to the membrane receptors, the receptors undergo conformational changes, which will permit the receptor to augment the exchange of GTP for GDP on a subunit of the trimeric G protein. The GTP-bound and activated G-protein subunits finally interact with their respective effectors to generate the appropriate signals (41). Here, we determined that the protective effect of α-MSH against DNA damage works in a MC1R-, XAB1-, and XPA-dependent manner. Thus, we postulated the signaling pathway of α-MSH–mediated DNA repair as shown in Supplementary Fig. S7. Following the binding of α-MSH, MC1R, which couples with a specific G protein, is activated and acts as a guanine nucleotide exchange factor for its respective G proteins. The α-MSH binding elicits a conformational change on MC1R, permitting it to augment the exchange of GTP for GDP on a subunit of the trimeric G protein. The XAB1 GTPase activity is then activated after the binding of GTP and eventually induces the nuclear translocation of the XPA protein to accelerate the clearance of photoproducts.

cAMP, a pivotal second messenger, regulates a diverse range of key cellular behaviors, including cell metabolism, cell growth and differentiation, apoptosis, and inflammatory responses. It is produced by an enzymatic reaction catalyzed by adenylyl cyclase. In eukaryotic cells, the effects of cAMP are mediated by two intracellular cAMP receptors, cAMP-dependent kinase (PKA), and Epac (42). Previous reports have shown that Epac proteins bind to cAMP with high affinity and activate the Ras superfamily small GTPase Rap1 and Rap2 (43). We showed that adenylyl cyclase is required in the activation of XAB1 GTPase activity needed for α-MSH–mediated DNA repair after UV irradiation. Our results also show that Epac, not PKA, is required for α-MSH–mediated DNA repair.

Cells use the NER system to eliminate DNA photoproducts. XPA is part of the core incision complex of the NER system (44). Here, we showed that α-MSH induces/augments XPA nuclear import and then activates the core incision complex in the NER system. We also found that the α-MSH–mediated translocation of XPA is modulated by XAB1. One group has shown that endogenous XPA protein is phosphorylated after UV irradiation in a lung adenocarcinoma cell line (45). However, we did not detect the phosphorylated XPA protein after UVB irradiation in keratinocytes nor any effect of α-MSH in the regulation of XPA phosphorylation. It is possible that the phosphorylated XPA is not as abundant in keratinocytes and therefore was below the detection threshold in our assay. It is also possible that XPA activity is modulated via different mechanisms in different cell types.

After UV-induced and p53-transactivated POMC/α-MSH is secreted from keratinocytes, it then binds to MC1R in both melanocytes and keratinocytes. Interestingly, distinct pathways downstream of MSH/MC1R become activated in these two cell types to reduce UV-induced DNA damages. In melanocytes, the α-MSH/MC1R complex activates the cAMP pathway and then induces melanin production. Melanin (eumelanin) functions as a physical barrier that scatters incident UV light, serving as a filter to reduce the penetration of UV light through the epidermis (46). In keratinocytes, the α-MSH/MC1R complex upregulates adenylyl cyclase activity and XAB1 GTPase activity inducing the nuclear translocation of XPA to repair the damaged DNA. In both the pigmentation induction and DNA repair activation processes, POMC/α-MSH functions as the crucial downstream target of the “guardian of genome” protein p53 (47).

Given the difficulty in curing melanoma and other types of skin cancers, there has been tremendous emphasis on their prevention. An obvious prevention strategy is to stimulate skin melanogenesis with small molecules. One class of pigment-inducing molecules is α-MSH and its analogues, which have shown major potential in skin cancer prevention. For example, an α-MSH analogue Melanotan has been shown to have sunless tanning capabilities and to reduce sun (UV) damage in preliminary studies and clinical trials (48). Another class of melanogenesis-promoting molecules are the cAMP agonists, such as forskolin (49). Forskolin activates adenylate cyclase and increases the intracellular levels of cAMP (49). Our previous data have shown that forskolin-induced pigmentation was protective against UV light–induced cutaneous DNA damage and tumorigenesis (4). One recent report showed that, independent of its effects on melanogenesis, forskolin protects keratinocytes from UVB-induced apoptosis and increases DNA repair (5). Forskolin is also able to mimic the antioxidative effect of α-MSH (50). Currently, there is no definitive data about the endogenous levels of α-MSH in human skin and its correlation with different skin types. Further pharmacokinetics study is required to identify the appropriate concentration of α-MSH required to reach the protective effect in the live tissue of human skin.

No potential conflicts of interest were disclosed.

We thank Drs.Guo Wei, Dali Liu, Huafeng Xie, and Jinyan Du for helpful comments.

Grant Support: NIH grant 1RO1CA137098, American Cancer Society grant RSG-09-022-01-CNE, American Cancer Society-Illinois Division grant 08-026, Elsa U. Pardee Foundation, and The Harry J. Lloyd Charitable Trust (R. Cui). R. Cui is an American Cancer Society Research Scholar.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.