Behind the Scene: Exploiting MC1R in Skin Cancer Risk and Prevention

Melanoma and non-melanoma skin cancers (NMSCs) are the most frequent cancers of the skin in white populations. An increased risk in the development of skin cancers has been associated with the combination of several environmental factors (i.e., ultraviolet exposure) and genetic background, including melanocortin-1 receptor (MC1R) status. In the last few years, advances in the diagnosis of skin cancers provided a great impact on clinical practice. Despite these advances, NMSCs are still the most common malignancy in humans and melanoma still shows a rising incidence and a poor prognosis when diagnosed at an advanced stage. Efforts are required to underlie the genetic and clinical heterogeneity of melanoma and NMSCs, leading to an optimization of the management of affected patients. The clinical implications of the impact of germline MC1R variants in melanoma and NMSCs’ risk, together with the additional risk conferred by somatic mutations in other peculiar genes, as well as the role of MC1R screening in skin cancers’ prevention will be addressed in the current review.


Introduction
Skin cancers represent the most frequent cancer, with 5 million of new cases each year [1]. The most common skin cancers include melanoma and non-melanoma skin cancers (NMSCs), among which basal cell carcinoma (BCC) and squamous cell carcinoma (SCC). For the last 10 years the incidence of melanoma has been continuously rising, together with that of NMSCs, although a precise estimation of the number of NMSCs is impaired by the fact that their reporting in cancer registries is not mandatory [1,2]. While BCC is usually confined to the skin, SCC and melanoma can be potentially metastatic, being difficult to treat and therefore presenting with a poor prognosis [3,4].
An increased risk of skin cancers has been associated with a combination of environmental agents, such as ultraviolet radiation (UVR), and genetic background. In particular, genes involved in pigmentary regulation, such as melanocortin-1 receptor (MC1R), are implicated in skin cancers' development [5,6]. The MC1R gene codifies for a G protein-coupled receptor (GPCRs), with a high affinity for α-melanocyte stimulating hormone (α-MSH). It is a highly polymorphic gene and it has been related to pigmentary as well as to non-pigmentary functions, including DNA repair.
In the last few years, there has been increasing knowledge about the MC1R functions and their clinical impact in dermato-oncology. The aim of this review is to provide an update on the impact of MC1R gene in melanoma and NMSCs' risk-together with the additional risk conferred by somatic mutation of other gene-as well as its role in the prevention of skin cancers.

MC1R Structure and Regulation
The MC1R gene (16q24.3, OMIM #155555) codes for seven transmembrane GPCRs of 317 amino acids, evolutionarily conserved [7,8], showing an extracellular N-terminus, with a glycosylation site, seven transmembrane segments, and an intracellular C-terminal extension including a palmitoylation site [9][10][11]. This receptor was first isolated from melanocytes, where its main physiological role in the ski has been shown [12]. MC1R shows a high affinity for the α-MSH, as well as for adrenocorticotropic hormone (ACTH) [9,10].
Additionally, a downregulation of cyclic-adenosine monophosphate (cAMP) signaling has been related to: an increase in phosphodiesterase 4D3 gene (PDE4D), a transcriptional target of cAMP via microphthalmia-associated transcription factor (MITF) [

MC1R and Pigmentation
MC1R has a pivotal role in pigmentation, although important non-pigmentary functions have also been identified [12].
The pathway leading to pigmentation has been widely characterized. It involves α-MSH binding to the MC1R, with a consequent increase in adenylyl-cyclase (AC) activity and rising of intracellular cAMP levels. cAMP activates protein kinase A (PKA), leading to the phosphorylation of the cAMP-response element binding protein CREB, which in turn activates the promoter of the MITF. The resultant event is the upregulation of tyrosinase (TYR) and Tyr-related proteins (TYRP1 and DCT) (Figure 1), switching pheomelanin to eumelanin synthesis [36,39-47] (Figure 1).

MC1R and Pigmentation
MC1R has a pivotal role in pigmentation, although important non-pigmentary functions have also been identified [12].
The pathway leading to pigmentation has been widely characterized. It involves α-MSH binding to the MC1R, with a consequent increase in adenylyl-cyclase (AC) activity and rising of intracellular cAMP levels. cAMP activates protein kinase A (PKA), leading to the phosphorylation of the cAMP-response element binding protein CREB, which in turn activates the promoter of the MITF. The resultant event is the upregulation of tyrosinase (TYR) and Tyr-related proteins (TYRP1 and DCT) (Figure 1), switching pheomelanin to eumelanin synthesis [36,39-47] (Figure 1). Binding of α-MSH on MC1R receptor activates adenylyl cyclase (AC) and stimulates cAMP production, which in turns induce the activation of several downstream effectors, including MITF transcription factor. MITF binds the MBOX on the promoters of tyrosinase (TYR), phosphoribosylanthranilate isomerase (TRP1) and dopachrome tautomerase (DCT) genes, leading to the expression of different enzymes involved in melanin biosynthesis. Melanin acts as a UV-protective shield in the epidermis. (Gα-β-γ proteins, CREB (cAMP response element binding protein), CBP (CREB-binding protein), CRE (cAMP response elements), c-KIT (tyrosine-protein kinase KIT), SH2 (Src homology 2), GEF (Guanine nucleotide Exchange Factor), MAPK/ERK (extracellular signal-regulated kinases)). This figure was created with BioRender.com.
Indeed, the functional impairment of MC1R downstream signaling is characterized by prevalent red/yellow pheomelanin. Pheomelanin has weak shielding capacity against UVR relative to eumelanin and has been shown to amplify UVA-induced reactive oxygen species (ROS). Thus, an increased ratio of photoprotective eumelanins to pro-oxidant pheomelanins provides an effective shield against mutagenic UVR [58]. Moreover, considering the link existing between α-MSH and PPAR-γ (Peroxisome Proliferator- The tyrosine-protein kinase c-KIT also plays a role in pigmentation [48-51]. Binding of the stem cell factor (SCF) to c-KIT induces sequential events [52,53], leading to recruitment of adaptor proteins containing a Src homology 2 (SH2) domain, which will associate with a guanine nucleotide exchange factor (GEF). The SH2/GEF complex activates RAS/RAF/MAPK/ERK cascade which, in turn, activates MITF [54][55][56][57] (Figure 1).
Indeed, the functional impairment of MC1R downstream signaling is characterized by prevalent red/yellow pheomelanin. Pheomelanin has weak shielding capacity against UVR relative to eumelanin and has been shown to amplify UVA-induced reactive oxygen species (ROS). Thus, an increased ratio of photoprotective eumelanins to pro-oxidant pheomelanins provides an effective shield against mutagenic UVR [58]. Moreover, considering the link existing between α-MSH and PPAR-γ (Peroxisome Proliferator-Activated Receptor γ), it has been shown that specific PPAR-γ modulators provide photoprotective effect in keratinocytes harboring MC1R-inactivating variants [59][60][61].

MC1R Polymorphisms
MC1R is a highly polymorphic gene with more than 200 variants described to date [9,[115][116][117]. Variant alleles including D84E, R142H, R151C, R160W and D294H have been defined as "R" or "RHC" alleles due to their strong association with the red hair color (RHC) phenotype [118]. The V60L, V92M and R163Q variants have a lower association with RHC phenotype and have been therefore designated as "r" alleles [9].
MC1R variants have been associated with a reduced receptor function, impairing the switch of melanin synthesis from eumelanin to the red-yellow pro-oxidant pheomelanin [115,118,119]. MC1R polymorphisms, both RHC and "r", generate hypomorphic proteins, leading to different degrees of cAMP pathway activation impairment, therefore leading to reduced pigmentary and non-pigmentary MC1R functions, as described in the previous sections [120,121]. Specifically, D84E, R151C and R160W polymorphisms have been related to a decreased cell surface expression [122][123][124], due to deficient anterograde trafficking or increased desensitization and internalization [18,125,126], while an impaired coupling has been reported for R142H and D294H alleles [122,123,127]. Only a marginal effect of the V92M substitution on cell-surface expression or ability to activate downstream signaling has been reported [122,123].

Clinical Impact of MC1R Polymorphisms: Hair and Skin Color and Non-Invasive Imaging Features
Considering the pivotal role of MC1R in pigmentation, MC1R status has been proven to have a key role on hair and skin color. Homozygotes, compound heterozygotes as well as heterozygotes for RHC MC1R alleles have been associated with red hair color [128,129].

MC1R Polymorphisms
MC1R is a highly polymorphic gene with more than 200 variants described to date [9,[115][116][117]. Variant alleles including D84E, R142H, R151C, R160W and D294H have been defined as "R" or "RHC" alleles due to their strong association with the red hair color (RHC) phenotype [118]. The V60L, V92M and R163Q variants have a lower association with RHC phenotype and have been therefore designated as "r" alleles [9].
MC1R variants have been associated with a reduced receptor function, impairing the switch of melanin synthesis from eumelanin to the red-yellow pro-oxidant pheomelanin [115,118,119]. MC1R polymorphisms, both RHC and "r", generate hypomorphic proteins, leading to different degrees of cAMP pathway activation impairment, therefore leading to reduced pigmentary and non-pigmentary MC1R functions, as described in the previous sections [120,121]. Specifically, D84E, R151C and R160W polymorphisms have been related to a decreased cell surface expression [122][123][124], due to deficient anterograde trafficking or increased desensitization and internalization [18,125,126], while an impaired coupling has been reported for R142H and D294H alleles [122,123,127]. Only a marginal effect of the V92M substitution on cell-surface expression or ability to activate downstream signaling has been reported [122,123].

Clinical Impact of MC1R Polymorphisms: Hair and Skin Color and Non-Invasive Imaging Features
Considering the pivotal role of MC1R in pigmentation, MC1R status has been proven to have a key role on hair and skin color. Homozygotes, compound heterozygotes as well as heterozygotes for RHC MC1R alleles have been associated with red hair color [128,129]. The V60L variant may act as a partially penetrant recessive allele. However, some indi-viduals carrying compound heterozygote/homozygote MC1R variants do not have red hair. A possible explanation might be that red hair color has also been related to mutations in other genes (i.e., POMC) [130]. A dosage effect of MC1R variants on hair, as well as skin color, should also be considered, being implicated in different shades of red hair in heterozygotes as compared to homozygotes/compound heterozygotes. There is also evidence for a heterozygote effect on beard hair color, skin type and freckling [128], although an association between MC1R RHC polymorphisms and freckles has been demonstrated to be independent of skin and hair color [131]. A dosage effect of MC1R variant alleles on sensitivity to UVR has also been described [28]. Accordingly, heterozygotes for one variant allele show an intermediate ability to tan after repeated sun exposure between those with two variant alleles (most likely to be red hair subjects) and those with none of the variants. Therefore, a high frequency of MC1R heterozygous allele carriers could influence the skin's response to UVR in most of the population who do not have red hair [132]. As a consequence, those who are homozygous/compound heterozygous for MC1R do not only have red hair, but also have pale skin, tan poorly and tend to burn on exposure to UVR, while subjects with pale skin who do not have red hair are more likely to be MC1R heterozygotes [128].
Non-invasive skin imaging performed with reflectance confocal microscopy (RCM) and optical coherence tomography (OCT), enabling in vivo evaluation of different layers of the skin, revealed a different dermal microenvironment in photoexposed skin of MC1R RHC variants carriers as compared to wild-type (WT) [133], suggesting a correlation between photoaging (aging related to UV exposure) in MC1R variant subjects and increased susceptibility to skin cancers [134]. Additionally, as revealed by clinical and daily routine non-invasive dermoscopy of nevi, MC1R status has an impact on nevus phenotype and RCM features. In detail, MC1R RHC variants carriers have a peculiar nevus phenotype, dermoscopically characterized by reduced structures and lower prevalence of atypical pigment network, visible vessels, dots and globules, and eccentric hyperpigmentation, associated with a high degree of skin freckling [135,136]. In addition, melanoma patients carrying MC1R variants as well as CDKN2A mutations, show clinically hypopigmented nevi and, at RCM, roundish cells infiltrating the dermo-epidermal junction [137].

MC1R and Melanoma Risk
Darker-pigmented subjects present a significantly higher risk of melanoma associated with MC1R variants [71,[147][148][149][150][151][152]. Individuals carrying just one MC1R variant have almost 40% increased risk of melanoma, whereas carriers of two or more MC1R variants have more than a double risk, as compared to WT subjects [124,153]. In particular, the association of MC1R RHC variant alleles D84E, R142H, R151C, R160W and D294H with a direct effect on melanoma risk has been confirmed by several studies and metaanalyses [71,[139][140][141]147,148,154,155]. However, melanoma in RHC individuals shows a significantly higher somatic mutational burden, as compared to melanoma patients without any RHC variants. Intriguingly, C > T and non-C > T were the most common mutations observed across all MC1R genotypes [156,157]. This might be related to a decreased protection against UVR damage in RHC carriers, or indicate that other mutational processes occur in melanocytes of these patients. Moreover, the number of MC1R variants also correlated positively with increased risk of melanoma development among individuals not showing the RHC phenotype [6,141]. A pooled analysis including 3830 melanoma cases and 2619 controls showed that the presence of any MC1R variant had a direct effect on melanoma, conferring a 60% higher risk to carriers versus non-carriers. Strikingly, considering the pigmentation-mediated effect of MC1R on melanoma risk prediction alone, it is smaller with any MC1R variant and each of the RHC and r variants [158]. Therefore, MC1R variants may partly mediate their effect through biological pathways that are independent of pigmentation and UVR [138][139][140][141]151].
A lower incidence and better survival rates for melanoma have been described in female subjects, as compared to males [159]. Interestingly, females carrying an RHC variant tended to exhibit significant lighter phototypes than males with the same MC1R genotypes, therefore contributing to different tanning ability between the two sexes.
Furthermore, MC1R variants have also been related to melanoma occurring in childhood and adolescents [160][161][162]. Interestingly, MC1R r variants were found to be more prevalent in childhood and adolescent melanoma than in adult ones, especially in patients aged 18 years or younger [163].

MC1R and NMSCs Risk
The most common NMSCs in fair-skinned populations are BCC and SCC [164]. UVR is the major environmental risk factor for NMSCs development [165], whereas fair skin and red hair are considered to be the most important phenotype risk factors [166]. Carriers of two MC1R variant alleles, mainly RHC variants, have a 2-to 3-fold increased risk of developing NMSCs, as compared to WT [131,153], also independently from phenotype [158].
In a study of 220 individuals (111 at high risk and 109 at low risk of BCC and SCC) in Queensland (area of high UVR), the prevalence of NMSCs was associated with the presence of MC1R RHC variant alleles R151C, R160W and D294H, whereas V60L, V92M and R163Q had minimal impact on BCC and SCC risk [167]. These findings were confirmed and extended in a case-control study of Dutch patients, showing the highest relative risks of NMSCs for D84E, H260P carriers, and slightly lower risks for R142H [131]. Data from the M-SKIP Project highlighted the association of MC1R variants and NMSCs development risk in populations living in different geographical areas, with a stronger role for darkerpigmented populations [70,124,131,143,168]. Interestingly, subjects with darker skin (skin types III and IV), carrying two MC1R variants, showed a lower risk of superficial multifocal BCC compared with MC1R variant carriers with lighter skin (skin types I and II), although the number of individuals in the analyzed subgroup was small [131].
The contribution of MC1R variants in the pathogenesis of each specific tumor type is not clear yet. Therefore, further investigation in functional studies focused on the carcinogenic mechanisms leading to BCC and SCC is needed [168].

MC1R Association with Melanoma Susceptible Genes
Germline MC1R variants may influence the mutational landscape of melanoma [146,156]. Despite the well-established impact of MC1R on skin cancer risk and development, the association of MC1R variants in combination with mutations in susceptible melanoma genes has not been clarified yet.
MC1R variants have been shown to increase the penetrance of CDKN2A mutations (observed risk over time for a mutation carrier), doubling the risk of melanoma development [124,154]. A stratified analysis of transmission of the R151C allele from parents to melanoma-positive offspring suggested that the contribution of the MC1R variant to the increased risk is independent of its effect on skin type [180]. Accordingly, a significant joint-effect of RHC variants (R163Q and D294H), considered either alone or in the presence of pigmentation and dysplastic nevi, influenced the penetrance of CDKN2A mutations in 20 French melanoma-prone families [181]. Helsing et al. reported that Norwegian melanoma patients showing both CDKN2A mutations and MC1R variants had an increased risk of melanoma when carrying D84E or R160W variants [182]. Additionally, carriers of A148T mutation of CDKN2A in association with non-synonymous MC1R variants (V60L, R151C, R160C and R163Q) have a 2-to 6-fold increased risk of melanoma [155,183]. Furthermore, germline carriers of the CDKN2A p16-Leiden deletion mutation in a large collection of Dutch families showed an increased risk of melanoma in carriers of MC1R variant alleles, with the R151C allele explaining most of this association [184].
In Queensland a CDKN2A mutation in association with MC1R variants have a raw penetrance of 84%, with a mean age at onset of 37.8 years when compared with family members who carry a CDKN2A mutation alone [179]. Accordingly, CDKN2A mutation carriers with MC1R variants had a significant lower median age at melanoma diagnosis than CDKN2A mutation carriers with no MC1R variants (37 years versus 47 years) [154,173,185]. Indeed, CDKN2A G101W mutation and MC1R variants carriers were younger at the first diagnosis with respect to WT multiple melanoma patients, showing hypopigmented nevi and roundish cells infiltrating the junction, suggesting an influence of CDKN2A mutation and MC1R variants in the development of dysplastic melanocytic lesions [137].

BRAF
Approximately 50% of melanomas harbor BRAF mutations. The association of MC1R variants with BRAF kinase proto-oncogene somatic mutations has been investigated in melanoma, showing different results among several populations.
An association between germline MC1R variants and somatic BRAF mutations was reported in tumors from United States and Italian populations. Carriers of at least one MC1R variant have a 5-to 15-fold increased risk confined only to BRAF+ melanomas, regardless the presence of chronic solar damage signs. On the contrary, no association with BRAFmelanomas was reported, suggesting that people carrying germline MC1R variants have a greater risk of developing a melanoma harboring a BRAF mutation without skin photodamage [186,187].
The association of MC1R variants (independently from the number of variants [188]) with somatic BRAF mutations has not been replicated in Italian [189,190], Spanish [191], German [192], Australian [193] and North Carolina [194] populations. These conflicting findings across different populations have been related to a different distribution of MC1R variants among study populations or a risk-modifying effect due to sun exposure.
Interestingly, a negative association between MC1R variants and BRAF mutations has been described for head/neck melanomas, suggesting a difference in the pathogenesis of melanomas located at different skin sites, head/neck or trunk, which could contribute to their divergent prognoses [195]. Additionally, a low frequency of somatic BRAF mutations in RHC and non-RHC MC1R carriers was restricted to nodular melanoma [192].

Other Genes
MC1R polymorphisms in association with other susceptible genes for melanoma have also been reported. Kosiniak-Kamysz et al. detected significant intermolecular epistasis effects among MC1R and TYR, SLC45A2 (solute carrier family 45 member 2) and vitamin D receptor gene (VDR), with MC1R RHC variants and TYR rs1393350 (G > A) showing the highest statistical significance [196]. Only three studies focused on somatic mutations in the TERT gene promoter, all of which reported moderate-to-strong positive associations with MC1R variants [189,190,197]. Several genetic interactions of MC1R in melanoma also include ASIP [198] and X-ray repair cross-complementing protein (XRCC) [70].

MC1R Association with NMSCs Susceptible Genes
The association of MC1R variants in combination with mutations in susceptible NM-SCs genes has not been investigated so far. The only evidence comes from Liboutet et al. who reported the P1315L mutation frequency in PTCH (Protein Patched Homolog 1, component of the hedgehog signaling pathway) not to be significantly different between BCC patients carrying a MC1R variant and those that do not carry one, suggesting an independent effect of both genes on BCC risk [199].
Interestingly, interactions between MC1R and melanoma susceptibility genes have been investigated, with inconsistent results [199].
However, future studies might be able to find potential correlations between different pathways leading to different great variability in NMSCs' aggressiveness, morphology and response to treatment.

Epigenetic Regulation of MC1R
Epigenetic factors such as DNA-methylation chromatin-remodeling events, as well as gene regulation through non-coding RNAs play an important role in the pathogenesis of skin cancers [200][201][202][203][204][205][206][207][208]. Interestingly, epigenetic regulation of MC1R expression in melanoma has been recently investigated [209]. A methylated CpG-island (CGI) has been identified on a MC1R region, proposed as a MC1R enhancer. This CGI has been shown to control MC1R expression, with a slight trend of increased methylation in melanomas showing homozygous RHC MC1R, as compared to WT and heterozygous tumors. Interestingly, unmethylated tumors had a significantly worse prognosis compared to methylated tumors, although the prognostic effect of MC1R CGI methylation has not been fully elucidated [209].
Epigenetic interactions have been recently identified in animal studies, showing some biological mechanisms underpinning their induction, such as dietary intake of cysteine, as well as miRNA targeting MC1R transcript, which may have an impact on receptor regulation [210,211].
Taken together, these data shed light on the complex regulation of MC1R and efforts should be made to fully elucidate epigenetic regulation of the receptor in humans.

MC1R and the Impact of Skin Cancer Genetic Testing
The well-established role of MC1R in skin cancer risk, which has been proven to be independent from skin phototype, highlights the importance of this gene in genetic testing and skin cancers' prediction. However, whether the feedback about genetic risk status may contribute to an increased skin cancer awareness, therefore leading to sun avoidance/protection and self-skin examination, and to a reduction of skin cancer risk is currently under debate [212].
A randomized controlled trial enrolling 73 patients with high risk of developing skin cancers was conducted to compare the impact of a strategy of CDKN2A/MC1R counseling, giving test results (intervention group), to a strategy of not offering genetic counseling and test results (usual care) on behaviors and skin cancer risk awareness. Patients enrolled were white, mainly females and college-educated. In particular, just one half of the patients were in the interventional group. This study, limited by the number of subjects enrolled and by the fact that just three patients were positive to CDKN2A/MC1R mutations/variants, did not show a significant impact of genetic counseling and test results on sun protection behaviors [213]. Another study, a randomized clinical trial by Hay et al., enrolling 499 patients, focused on the interest in and uptake of MC1R testing in the general population, after offering information about advantages and disadvantages of the test. Most of the people enrolled were non-Hispanic white, and had a high school diploma or less. Just a few patients were at a high risk for skin cancers, although more than one half of patients experienced sunburns. Interestingly, college-educated non-Hispanic white were significantly more prone to read information about MC1R testing while subjects experiencing sunburns were significantly more likely to request the test [214]. Importantly, a recent study showed that people with lower health literacy skills or education may need support to understand genetic test results, while higher skills were related to reduced distress after receiving the results of MC1R testing [215].
This result underlines the importance of personalized education for the correct communication with patients with different skills, in order to avoid the distress that might be related to the knowledge of increased susceptibility to skin cancers based on genetic information. However, the impact of MC1R testing on skin cancer awareness and sun avoidance behavior has not been established yet.
Interestingly, previous studies explored the role of MC1R as a prognostic marker in metastatic melanoma and as a potential approach for target treatments in skin cancers. A significant correlation between MC1R variants and worse outcomes (overall response rate and progression-free survival) in BRAF-mutated metastatic melanoma patients treated with target therapy was observed, due to the interactions of MC1R with other pathways [216].
Additionally, considering that impaired function of MC1R has been related to an increased susceptibility to skin cancers, MC1R agonists and antagonists might be employed as a potential therapeutic approach [12,217]. Accordingly, in vitro and animal studies have shown that forskolin, through increasing cAMP levels, induced an improvement in NER function and DNA repair [108,218,219]. Furthermore, regulation of palmitoylation, which has been shown to be reduced in RHC MC1R, has been proven to reduce melanoma risk in in vitro and animal models [11,19].
However, MC1R has not been employed as a target treatment for skin cancers in humans so far. Currently, an α-MSH analogue is employed in clinical practice for treating photosensitivity in patients with erythropoietic protoporphyria [220].
Based on the information currently available in the literature, future studies are needed in order to provide data concerning the role of personalized genomic risk, including MC1R, with potential clinical impact in terms of early detection, treatment and preventive strategies.

Conclusions
As knowledge is expanding very rapidly, future directions should be addressed to determine the biological mechanisms underlying non-pigmentary MC1R functions, evaluate the gene-gene and gene-environment interactions, and to incorporate MC1R variants into melanoma and NMSCs risk prediction models and test their effect on motivating risk-reducing behaviors as a cancer prevention strategy. Additionally, non-invasive skin imaging evaluation, together with genetic studies, might improve the recognition of early skin variations preceding skin tumors' development as well as melanoma and NMSCs' identification and early tumor diagnosis, for a patient-tailored management protocol.