A polymorphic p53 response element in KIT ligand influences cancer risk and has undergone natural selection.

The ability of p53 to regulate transcription is crucial for tumor suppression and implies that inherited polymorphisms in functional p53-binding sites could influence cancer. Here, we identify a polymorphic p53 responsive element and demonstrate its influence on cancer risk using genome-wide data sets of cancer susceptibility loci, genetic variation, p53 occupancy, and p53-binding sites. We uncover a single-nucleotide polymorphism (SNP) in a functional p53-binding site and establish its influence on the ability of p53 to bind to and regulate transcription of the KITLG gene. The SNP resides in KITLG and associates with one of the largest risks identified among cancer genome-wide association studies. We establish that the SNP has undergone positive selection throughout evolution, signifying a selective benefit, but go on to show that similar SNPs are rare in the genome due to negative selection, indicating that polymorphisms in p53-binding sites are primarily detrimental to humans.

Murine dorsal hair type is genetically determined by polymorphisms in candidate genes that influence BMP and WNT signalling.

Mouse dorsal coat hair types, guard, awl, auchene and zigzag, develop in three consecutive waves. To date, it is unclear if these hair types are determined genetically through expression of specific factors or can change based on their mesenchymal environment. We undertook a novel approach to this question by studying individual hair type in 67 Collaborative Cross (CC) mouse lines and found significant variation in the proportion of each type between strains. Variation in the proportion of zigzag, awl and auchene, but not guard hair, was largely due to germline genetic variation. We utilised this variation to map a quantitative trait locus (QTL) on chromosome 12 that appears to influence a decision point switch controlling the propensity for either second (awl and auchene) or third wave (zigzag) hairs to develop. This locus contains two strong candidates, Sostdc1 and Twist1, each of which carry several ENCODE regulatory variants, specific to the causal allele, that can influence gene expression, are expressed in the developing hair follicle, and have been previously reported to be involved in regulating human and murine hair behaviour, but not hair subtype determination. Both of these genes are likely to play a part in hair type determination via regulation of BMP and/or WNT signalling.

Further assessment of exome-wide UVR footprints in melanoma and their possible relevance.

C>T substitutions at dipyrimidine sites dominate the melanoma genome. We recently analyzed the exomes of spontaneous and neonatal UVR-induced murine melanomas, noting a dramatic change in the genomic footprint at C>T substitutions in the latter. Here we re-analyzed published exome-wide footprints in human melanomas stratified in terms of likely previous sun exposure. Acral and mucosal melanomas were heterogeneous in terms of base substitution types, but most C>Ts occurred in the context of 3'G, probably resulting from spontaneous deamination of the cytosine. C>Ts in sun-exposed melanomas were statistically different from acral/mucosal lesions only in preferring an adjacent 5'T and 3'C. Pyrimidine dimer adducts can form between any pyrimidine (TT, TC, CT, CC). Hence in melanoma C>Ts are overwhelmingly induced at TC or CC photoproducts, or, there are peculiarities in DNA repair that favor the mutation of cytosines with these two pyrimidines adjacent. If melanoma UVR footprints at C>Ts reflect a specific dimer type (eg, 6-4 photoproduct or cyclobutane pyrimidine dimer), these could be removed post UVR, for instance using photolyases, to potentially reduce melanoma risk. If specific modes of DNA repair and/or replication cause these footprints, methodically downregulating selected DNA polymerases in UVR-induced animal models of melanoma, combined with exome sequencing, could begin to assess this. Finally, a preponderance of TpCpC as opposed to NpCpG at C>Ts exome-wide is likely to be a good indicator of whether a melanoma has incurred even a small amount of sun damage. This information will assist epidemiological studies in predicting individual levels of sun exposure.

Hair follicle melanocyte precursors are awoken by ultraviolet radiation via a cell extrinsic mechanism.

Melanocyte stem cells (MCSCs) in the upper portion of the hair follicle periodically supply melanocytes (MCs) that migrate downward into the hair bulb during anagen, the growth phase of the hair cycle. However MCs can also migrate upwards. We previously observed an increase in epidermal MC density in the mouse epidermis after a single ultraviolet radiation (UVR) exposure in neonatal, but not adult mice. To better understand MCSC activation by UVR we methodically studied the response of MCs to narrow band UVB (since UVA does not invoke this response) exposure in neonatal mice, and in adults at different stages of the hair cycle. We found that a single exposure of adult mice did not induce activation of MCSCs, in any stage of the hair cycle. When adult mice MCSCs were isolated in telogen, multiple UVB exposures resulted in their activation and production of daughter cells, which migrated upwards to the epidermis. Importantly, the MCSCs produced new progeny without themselves having incurred DNA damage after UVB exposure. This, together with examination of MC localisation in the skin of mice overexpressing stem cell factor in their keratinocytes, leads us to conclude that MCSC activation by UVB is driven via paracrine production of either SCF and/or other keratinocyte cytokines. We re-examined the increase in epidermal MC density in neonatal mouse skin. This effect was much more profound after only a single exposure than that of even multiple exposures to adult skin, and we show that in this setting also, the epidermal MCs mostly derive from activation of MC precursors in the upper hair follicle, and most likely via a cell extrinsic mechanism. Hence, although adaptive changes in the skin induced by repetitive UVB exposures are necessary in adult mice, in both the adult and neonatal context the division and migration upwards of follicular MCSCs is the major mode by which epidermal MC numbers increase after UVR exposure.

A blueprint for staging of murine melanocytic lesions based on the Cdk4 ( R24C/R24C ) ::Tyr- NRAS ( Q ) ( 61K ) model.

It has been shown that gene mutations which drive the development of malignant melanoma (MM) in humans also lead to emergence of MM when engineered mice. However, little attention has been paid to the clinical and histopathological features of melanocytic lesions and their natural history in a given mouse model. This knowledge is crucial to enable us to understand how engineered mutations influence the initiation and evolution of melanocytic lesions, and/or for the use of mice as a preclinical model to test specific treatments. We recently reported the development of melanocytic proliferations along the spectrum of naevi to MM in a Cdk4 ( R24C/R24C ) ::Tyr- NRAS ( Q ) ( 61K ) mouse model. In this study, we followed the development of lesions over time using digital photography and dermoscopy with the aim to correlate the clinical and histopathological features of lesions developing in this model. We identified two types of lesions. The first are slow-growing dermal MMs that emanate from dermal naevi. The second did not emanate from naevi, grew rapidly, and appeared to be solely confined to the subcutaneous fat. We present a simple staging system for the MMs that progress from naevi, based on depth of extension into the dermis and subcutis. This represents a blueprint for documentation and follow-up of MMs in the live animal, which is critical for the proper use of murine melanoma models.

hSSB2 (NABP1) is required for the recruitment of RPA during the cellular response to DNA UV damage.

Maintenance of genomic stability is critical to prevent diseases such as cancer. As such, eukaryotic cells have multiple pathways to efficiently detect, signal and repair DNA damage. One common form of exogenous DNA damage comes from ultraviolet B (UVB) radiation. UVB generates cyclobutane pyrimidine dimers (CPD) that must be rapidly detected and repaired to maintain the genetic code. The nucleotide excision repair (NER) pathway is the main repair system for this type of DNA damage. Here, we determined the role of the human Single-Stranded DNA Binding protein 2, hSSB2, in the response to UVB exposure. We demonstrate that hSSB2 levels increase in vitro and in vivo after UVB irradiation and that hSSB2 rapidly binds to chromatin. Depletion of hSSB2 results in significantly decreased Replication Protein A (RPA32) phosphorylation and impaired RPA32 localisation to the site of UV-induced DNA damage. Delayed recruitment of NER protein Xeroderma Pigmentosum group C (XPC) was also observed, leading to increased cellular sensitivity to UVB. Finally, hSSB2 was shown to have affinity for single-strand DNA containing a single CPD and for duplex DNA with a two-base mismatch mimicking a CPD moiety. Altogether our data demonstrate that hSSB2 is involved in the cellular response to UV exposure.

Keratinocyte Cytokine Networks Associated with Human Melanocytic Nevus Development.

Melanocytes can group together in nevi, commonly thought to form because of intrinsic somatic mutations involving MAPK pathway activation. However, the role of the microenvironment, in particular keratinocytes, in nevogenesis is rarely studied. Melanocytes proliferate during the hair follicle growth phase and in some basal cell carcinomas, allowing us to construct keratinocyte gene expression clusters correlated with melanocyte activation. We asked whether such correlations are evident in the more subtle context of regulation of melanocyte behavior in normal skin. We considered genes which, when mutated in keratinocytes in mice, lead to nevogenesis. Across the human GTEx normal skin database, their expression was correlated with that of keratinocyte cytokines KITLG, HGF, FGF2, EDN1, and melanocyte markers. These cytokines have pleiotropic effects on melanocyte-specific and pigmentation genes and also influence mast cell gene expression. We show five classes of keratinocyte genes that, via germline genetic variation, influence melanocyte activity. These include genes involved in SHH signaling, structural keratins, ribosomal biogenesis, and stem cell governance. In agreement with the finding of KITLG linked to nevogenesis in human genome-wide association studies, we provide evidence that specific keratinocyte cytokines are components of networks that may drive or exacerbate nevus development.

Different genetic mechanisms mediate spontaneous versus UVR-induced malignant melanoma.

Genetic variation conferring resistance and susceptibility to carcinogen-induced tumorigenesis is frequently studied in mice. We have now turned this idea to melanoma using the collaborative cross (CC), a resource of mouse strains designed to discover genes for complex diseases. We studied melanoma-prone transgenic progeny across seventy CC genetic backgrounds. We mapped a strong quantitative trait locus for rapid onset spontaneous melanoma onset to Prkdc, a gene involved in detection and repair of DNA damage. In contrast, rapid onset UVR-induced melanoma was linked to the ribosomal subunit gene Rrp15. Ribosome biogenesis was upregulated in skin shortly after UVR exposure. Mechanistically, variation in the 'usual suspects' by which UVR may exacerbate melanoma, defective DNA repair, melanocyte proliferation, or inflammatory cell infiltration, did not explain melanoma susceptibility or resistance across the CC. Instead, events occurring soon after exposure, such as dysregulation of ribosome function, which alters many aspects of cellular metabolism, may be important.

Keratinocyte Sonic Hedgehog Upregulation Drives the Development of Giant Congenital Nevi via Paracrine Endothelin-1 Secretion.

Giant congenital nevi are associated with clinical complications such as neurocutaneous melanosis and melanoma. Virtually nothing is known about why some individuals develop these lesions. We previously identified the sonic hedgehog (Shh) pathway regulator Cdon as a candidate nevus modifier gene. Here we validate this by studying Cdon knockout mice, and go on to establishing the mechanism by which Shh exacerbates nevogenesis. Cdon knockout mice develop blue nevi without the need for somatic melanocyte oncogenic mutation. In a mouse model carrying melanocyte NRASQ61K, we found that strain backgrounds that carry genetic variants that cause increased keratinocyte Shh pathway activity, as measured by Gli1 and Gli2 expression, develop giant congenital nevi. Shh components are also active adjacent to human congenital nevi. Mechanistically, this exacerbation of nevogenesis is driven via the release of the melanocyte mitogen endothelin-1 from keratinocytes. We then suppressed nevus development in mice using Shh and endothelin antagonists. Our work suggests an aspect of nevus development whereby keratinocyte cytokines such as endothelin-1 can exacerbate nevogenesis, and provides potential therapeutic approaches for giant congenital nevi. Furthermore, it highlights the notion that germline genetic variation, in addition to somatic melanocyte mutation, can strongly influence the histopathological features of melanocytic nevi.

Genetic variation in IRF4 expression modulates growth characteristics, tyrosinase expression and interferon-gamma response in melanocytic cells.

A SNP within intron4 of the interferon regulatory factor4 (IRF4) gene, rs12203592*C/T, has been independently associated with pigmentation and age-specific effects on naevus count in European-derived populations. We have characterized the cis-regulatory activity of this intronic region and using human foreskin-derived melanoblast strains, we have explored the correlation between IRF4 rs12203592 homozygous C/C and T/T genotypes with TYR enzyme activity, supporting its association with pigmentation traits. Further, higher IRF4 protein levels directed by the rs12203592*C allele were associated with increased basal proliferation but decreased cell viability following UVR, an etiological factor in melanoma development. Since UVR, and accompanying IFNγ-mediated inflammatory response, is associated with melanomagenesis, we evaluated its effects in the context of IRF4 status. Manipulation of IRF4 levels followed by IFNγ treatment revealed a subset of chemokines and immuno-evasive molecules that are sensitive to IRF4 expression level and genotype including CTLA4 and PD-L1.

A mutation in the Cdon gene potentiates congenital nevus development mediated by NRAS(Q61K).

Congenital nevi develop before birth and sometimes cover large areas of the body. They are presumed to arise from the acquisition of a gene mutation in an embryonic melanocyte that becomes trapped in the dermis during development. Mice bearing the Cdk4(R24C) ::Tyr-NRAS(Q) (61K) transgenes develop congenital nevus-like lesions by post-natal day 10, from melanocytes escaping the confines of hair follicles. We interbred these mice with the collaborative cross (CC), a resource that enables identification of modifier genes for complex diseases (those where multiple genes are involved). We examined variation in nevus cell density in 66 CC strains and mapped a large-effect quantitative trait locus (QTL) controlling nevus cell density to murine chromosome 9. The best candidate for a gene that exacerbates congenital nevus development in the context of an NRAS mutation is Cdon, a positive regulator of sonic hedgehog (Shh) that is expressed mainly in keratinocytes.

Microarray expression profiling in melanoma reveals a BRAF mutation signature.

We have used microarray gene expression profiling and machine learning to predict the presence of BRAF mutations in a panel of 61 melanoma cell lines. The BRAF gene was found to be mutated in 42 samples (69%) and intragenic mutations of the NRAS gene were detected in seven samples (11%). No cell line carried mutations of both genes. Using support vector machines, we have built a classifier that differentiates between melanoma cell lines based on BRAF mutation status. As few as 83 genes are able to discriminate between BRAF mutant and BRAF wild-type samples with clear separation observed using hierarchical clustering. Multidimensional scaling was used to visualize the relationship between a BRAF mutation signature and that of a generalized mitogen-activated protein kinase (MAPK) activation (either BRAF or NRAS mutation) in the context of the discriminating gene list. We observed that samples carrying NRAS mutations lie somewhere between those with or without BRAF mutations. These observations suggest that there are gene-specific mutation signals in addition to a common MAPK activation that result from the pleiotropic effects of either BRAF or NRAS on other signaling pathways, leading to measurably different transcriptional changes.

Mouse models for actinic keratosis and squamous cell carcinoma.

This manuscript focuses on the use of mice to study the genetics and biology of cutaneous squamous cell carcinoma (SCC). Mice develop actinic keratosis-like lesions and SCC resembling those seen in humans. As an animal model, the mouse provides great experimental flexibility and has been useful in investigating aspects of the genetics and biology of SCC that are difficult to study in humans. We discuss the pros and cons of the various murine models available. How well mouse pathology in general mimics human disease remains an open question due to the vast differences in animal strain backgrounds and the fact that only one strain is typically tested in any particular experiment. Nonetheless, the murine epidermis is thinner than the human epidermis, and this must be kept in mind when making inferences from mechanistic data obtained with mice. We outline new strategies for non-biased screens to discover genes driving SCC progression. Such work has revealed a very complex interactive molecular network, and as with other complex diseases, the picture is being pieced together using systems biology strategies to which mouse tumour models are amenable. Such approaches do not focus on single genes or proteins but try to integrate the complex interactions of many types of genetic and biological information.

Clinicopathological characterization of mouse models of melanoma.

Mouse models of melanoma have proven invaluable in the delineation of key molecular events involved in disease progression in humans and provide potential preclinical models for therapeutic testing (Damsky and Bosenberg, Pigment Cell Melanoma Res 25(4):404-405, 2012; Walker et al., Pigment Cell Melanoma Res 24(6):1158-1176, 2011). Here we concentrate on the clinicopathological analysis of melanocytic tumors.

Pathways to melanoma development: lessons from the mouse.

Because of subtle differences between mouse and human skin, mice have traditionally not been an ideal model to study melanoma development. Understanding of the molecular mechanisms of melanoma predisposition, however, has been greatly improved by modeling various pathway defects in the mouse. This review analyzes the latest developments in mouse models of melanoma, and summarizes what these may indicate about the development of this neoplasm in humans. Mutations of genes involved in human melanoma have been recapitulated with some unexpected results, particularly with respect to the role of the two transcripts (Ink4a and Arf) encoded by the Cdkn2a locus. Both the Ink4a/pRb and Arf/p53 pathways are involved in melanoma development in mice, and possible mechanisms of cross-talk between the two pathways are discussed. We also know from mouse models that Ras/mitogen-activated protein kinase pathway activation is very important in melanoma development, either through direct activation of Ras (e.g., Hras G12V), or via activation of Ras-effector pathways by other oncogenes (e.g., Ret, Hgf/Sf). Ras can cooperate with the Arf/p53 pathway, and probably the Ink4a/Rb pathway, to induce melanoma. These three growth regulation pathways (Ink4a/pRb, Arf/p53, and Ras/mitogen-activated protein kinase) seem to represent three major "axes" of melanoma development in mice. Finally, we summarize experiments using genetically modified mice that have given indications of the intensity and timing of ultraviolet radiation exposure that may be most responsible for melanoma development.

Differential effects of ultraviolet irradiation in neonatal versus adult mice are not explained by defective macrophage or neutrophil infiltration.

Epidemiological studies suggest that ultraviolet B exposure (UVR) during childhood is the most important environmental risk factor for melanoma. In accordance, neonatal, but not adult, UVR exacerbates melanoma incidence in mouse models. The inability of neonates, as opposed to adults, to mount a proper neutrophil inflammatory response in the skin upon UVR exposure has been one of the driving hypotheses explaining this observation for the past decade. However, this aspect remains controversial. Here, we evaluated the UVR-induced inflammatory response in neonatal versus adult mice. In neonates, a significant neutrophil infiltration could be identified and quantified using three different antibodies by flow cytometry or immunohistochemistry. On day 1 after UVR, neutrophils were increased by 84-fold and on day 4 macrophages increased by 37-fold compared with nonexposed age-matched skin. When compared with adults, neonatal skin harbored a higher proportion of neutrophils in the myeloid compartment without significant differences in absolute counts. This response was reproduced with different kinetics in C57Bl/6 and FVB mice with a more rapid attenuation of neutrophil counts in the latter. Overall, our results suggest that the greatly increased sensitivity to melanomagenesis in neonates does not result from their incompetence in terms of myeloid inflammatory response to UVR.

UVB-induced melanocyte proliferation in neonatal mice driven by CCR2-independent recruitment of Ly6c(low)MHCII(hi) macrophages.

Intermittent sunburns, particularly in childhood, are the strongest environmental risk factor for malignant melanoma (MM). In mice, a single neonatal UVR exposure induces MM, whereas chronic doses to adult mice do not. Neonatal UVR alters melanocyte migration dynamics by inducing their movement upward out of hair follicles into the epidermis. UVR is known to induce inflammation and recruitment of macrophages into the skin. In this study, we have used a liposomal clodronate strategy to deplete macrophages at the time of neonatal UVR, and have shown functionally that this reduces the melanocyte proliferative response. This effect was not reproduced by depletion of CD11c-expressing populations of dendritic cells. On the basis of epidermal expression array data at various time points after UVR, we selected mouse strains defective in various aspects of macrophage recruitment, activation, and effector functions, and measured their melanocyte UVR response. We identified Ly6c(low)MHCII(hi) macrophages as the major population promoting the melanocyte response across multiple strains. The activity of this subpopulation was CCR2 (C-C chemokine receptor type 2) independent and partly IL-17 dependent. By helping induce this effect, the infiltration of specific macrophage subpopulations after sunburn may be a factor in increasing the risk of subsequent neoplastic transformation of melanocytes.

Plasticity of melanoma in vivo: murine lesions resulting from Trp53, but not Cdk4 or Arf deregulation, display neural transdifferentiation.

We previously noted that melanomas developing in Cdk4(R24C/R24C) ::Tyr-NRAS, Arf(-/-) ::Tyr-NRAS and Trp53(F/F) ::Tyr-Cre(ER)::Tyr-NRAS mice exhibited differences in behaviour in vivo. We investigated this phenomenon using global gene expression profiling of lesions from the respective genotypes. While those from the Cdk4- and Arf-mutant mice exhibited similar profiles, the Trp53(F/F) ::Tyr-Cre(ER)::Tyr-NRAS melanomas were strikingly different, showing relative down-regulation of melanocyte-related genes, and up-regulation of genes related to neural differentiation. Specifically, they highly expressed genes representative of the myelin-producing peripheral oligodendrite (Schwann cell) lineage, although histopathologically the lesions did not exhibit the classical features of schwannoma. As Schwann cell precursors can be a cellular origin of melanocytes, it is unsurprising that plasticity with respect to melanocyte-neural differentiation can occur in melanoma. What is surprising is the genotype proclivity. Comparison of gene expression signatures revealed that melanomas from the Trp53-mutant mice show significant similarities with a subset of aggressive human melanomas with relatively low levels of MITF.