Oncogenic NRAS Cooperates with p53 Loss to Generate Melanoma in Zebrafish

doi:10.1089/zeb.2009.0606

Zebrafish. 2009 December; 6(4): 397–404.

doi: 10.1089/zeb.2009.0606

PMCID: PMC2943216

NIHMSID: NIHMS231154

Oncogenic NRAS Cooperates with p53 Loss to Generate Melanoma in Zebrafish

Michael Dovey, Richard Mark White, and Leonard I. Zon

Stem Cell Program and Hematology/Oncology, Children's Hospital; Dana-Farber Cancer Institute, Howard Hughes Medical Institute, Harvard Stem Cell Institute, Harvard Medical School, Boston, Massachusetts.

Corresponding author.

Address correspondence to: Leonard I. Zon, M.D., Children's Hospital Boston, Karp Family Research Building, Room 7211, 1 Blackfan Circle, Boston, MA 02115. E-mail:Email: zon/at/enders.tch.harvard.edu

This article has been cited by other articles in PMC.

Abstract

NRAS mutations are a common oncogenic event in skin cancer, occurring frequently in congenital nevi and malignant melanoma. To study the role of NRAS in zebrafish, a transgenic approach was applied to generate fish that express human oncogenic NRAS^Q61K under the control of the melanocyte-restricted mitfa promoter. By screening the progeny of the injected animals, two strains stably expressing the NRAS transgene were identified: Tg(mitfa:EGFP:NRAS^Q61K)₁ and Tg(mitfa:EGFP:NRAS^Q61K)₂. Stable expression of this transgene results in hyperpigmented fish displaying a complete ablation of the normal pigment pattern. Although oncogenic NRAS expression alone was found to be insufficient to promote tumor formation, loss of functional p53 was found to collaborate with NRAS expression in the genesis of melanoma. The tumors derived from these animals are variably pigmented and closely resemble human melanoma. Underscoring the pathological similarities between these tumors and human disease and suggesting that common pathways are similar in these models and human disease, gene set enrichment analysis performed on microarray data found that the upregulated genes from zebrafish melanomas are highly enriched in human tumor samples. This work characterizes two zebrafish melanoma models that will be useful tools for the study of melanoma pathogenesis.

Introduction

Melanoma is the cancer of the pigment-producing cells, melanocytes. Although melanoma only accounts for 5% of all reported cases of skin cancer in the United States, it is the leading cause of mortality among the skin cancer patients.¹ The propensity of melanomas to quickly metastasize often results in poor patient prognosis and worse survival. Aside from the modest success of interferon therapy, no chemotherapeutic regiment has been shown to be widely efficacious.² Because of the aggressive nature of melanoma and the need for improved treatments, the molecular mechanism of melanoma has been heavily studied.

The primary molecular basis underlying melanoma pathogenesis has been described. The aberrant activation of mitogen-activated protein kinase (MAPK) signaling is the most common oncogenic event in melanoma. Activating mutations in either BRAF or NRAS are present in almost all (95%) human melanomas.^3,4 BRAF mutations are often accompanied by losses in the tumor suppressor gene PTEN.⁵ NRAS is mutated in nearly 33% of the primary human melanoma and as many as 26% of the metastatic disease.⁶ The most common variants affect codon 61, which accounts for nearly all RAS mutations in melanoma.⁷

Even though constitutively activated MAPK signaling has been shown to contribute to the cellular growth, mutations in NRAS and BRAF are, alone, not sufficient for tumorigenesis.^8–10 This is exemplified by the observation that oncogenic mutations in BRAF are commonly found in sporadic nevi that do not develop into tumors.^11,12 That NRAS mutations are rarely seen in dysplastic nevi may suggest a distinct path to cancer from BRAF.¹³

Experimentally, oncogenic BRAF or HRAS is capable of contributing to melanoma, but only in the context of cooperating loss-of-function mutations in p53, INK4, and/or ARF tumor suppressors.^14,15 The loss of these genes has been shown to further cancer progression and lead to the disruption of Rb signaling, either directly or through MYC amplification. Additional genetic evidence from mice suggests that ARF loss contributes to melanoma progression.¹⁶ Although mutations in p53 are rare in human melanoma, this experimental evidence suggests a role for this pathway in melanoma. The fact that germline CDKN2A mutations have been identified in cases of familial melanoma further underscores the importance of these pathways in melanoma progression.¹⁷ Recent studies in mouse have also shown that NRAS^Q61K is capable of promoting melanoma formation and cooperates with INK4A loss to accelerate tumor onset.¹⁸

Our laboratory has previously described a cancer model in zebrafish in which the human oncogenic BRAF^V600E, expressed in a melanocyte-restricted manner, cooperates with p53 loss in the genesis of melanoma.¹⁴ In addition to providing a useful platform for further experimentation, these studies support the previous observations that p53 signaling plays an important role in BRAF-driven melanoma progression. We sought to complement these studies by analyzing the effect of oncogenic NRAS on zebrafish melanocytes. To that end, we generated two NRAS-driven melanoma models in zebrafish. Human oncogenic NRAS^Q61K was expressed under the melanophore-specific mitfa promoter and N-terminally tagged with EGFP. Two stable transgenic zebrafish lines were identified and used for analysis: NRAS1, which represents Tg(mitfa:EGFP:NRAS^Q61K)₁, and NRAS2, which represents Tg(mitfa:EGFP:NRAS^Q61K)₂. Both transgenic lines have severe disruptions in normal pigment patterning, with NRAS1 animals having a more severe defect. In the absence of normal p53, these strains develop variably pigmented tumors that resemble very closely the pathology of human melanoma. Gene set enrichment analysis (GSEA) of microarrays further confirmed the molecular similarity between our zebrafish tumors and human melanomas. In summary, NRAS is a potent oncogene in zebrafish melanocytes, and melanomas derived from this model closely mimic the human disease.

Methods

Cloning

Full-length human NRAS was amplified from clone IOH1768 (Invitrogen, Carlsbad, CA). The clone was first mutagenized to the 61K variant by site-directed mutagenesis performed using 5′-ACTGAATACAGCTGGAAAAGAAGAGTACAGTGCCA-3′ and 5′-TGGCACTGTACTCT TCTTTTCCAGCTGTATCCAGT-3′ primers with the Quick Change site-directed mutagenesis kit (Stratagene, La Jolla, CA). The mutant expressed sequence tag (EST) was then amplified with linker-containing primers 5′-AAAGATCTATGACTGAGTACA AACTGGTGG-3′ and 5′-AAGGGCCCTACATCAC CACACATGGCAATC-3′ and cloned in frame into the pEGFP-C1 plasmid (Invitrogen) with BglII and ApaI. About 1

kb of the zebrafish mitfa promoter was polymerase chain reaction (PCR) amplified from the mitfa:BRAF^V600E construct with primers 5′-AAGCTAGCGTATATGCACATGCTACT TTGT-3′ and 5′-AAAGCGCTTGTTCAACTATGTGTTAGCTTC-3′. The promoter was then cloned into the pEGFP:NRAS^Q61K vector with NheI and AfeI. The full-length transgene was sequenced for accuracy.

Transgenesis

The mitfa:EGFP:NRAS^Q61K transgene was excised from the plasmid backbone with NheI and MluI, separated by gel electrophoresis, and purified. Naked DNA was injected at a concentration of 100

ng/μL into one-cell stage embryos. To determine germline transgenesis, F0 fish were in-crossed and the F1 progeny screened by PCR with 5′-CACATGAAGCAGCACGACTT-3′ and 5′-ACGTAAACGGCCACAAG TTC-3′.

Copy number analysis

Transgenic copy number was determined as described.¹⁹ DNA from whole transgenic or wild-type fish was isolated by phenol–chloroform extraction. Linear plasmid was calculated to have 147,400 transgene copies per picogram of plasmid DNA and used in limiting dilution to generate a standard curve for copy number determination. The zebrafish genome was assumed to have a size of 1.5

Gb per haploid genome. Sybr Green (Invitrogen) quantitative PCR-directed mitfa promoter was used for comparisons between wild-type, NRAS1, and NRAS2 fish. At least three technical replicates were performed, as were two biological replicates, which were averaged. The primers used were directed against the mitfa promoter: 5′-GCAGGACTTCAAATGACAAACACG-3′ and 5′-TGAGGGGCAGGAGTTACT GATG-3′.

Real time PCR

RNA was extracted from NRAS1 and NRAS2 scales. cDNA was generated, and transgene expression was measured by quantitative PCR with Sybr Green Two-Step pRT-PCR Kit (Invitrogen) with at least three technical replicates. Beta-actin was used as a control. The EGFP primers 5′-AAGTCGTGCTGCTTCAT GTG-3′ and 5′-ACGTAAA CGGCCACAAGTTC-3′ were used to detect transgene expression. Actin was detected with 5′-GCTGTTTTCCCCCTCCATTGTT-3′ and 5′-TCCCATGCCAACCATCACT-3′.

Pathology

Fish were euthanized and fixed in 4% paraformaldehyde overnight. They were then decalcified in 0.5 M ethylenediaminetetraacetic acid before paraffin embedding and sectioning. Staining and immunohistochemistry were done according to the standard techniques by the Brigham and Women's Pathology Core. RNA in situ probe for mitfa was a generous gift from Rodney Stewart in A. Thomas Look's Laboratory.²⁰

Transplantation

Briefly, tumors were dissected from anesthetized animals and minced. Tumors were maintained in 0.9

phosphate-buffered saline supplemented with 5% fetal calf serum and passed through a 40

μm mesh filter before sorting. Cell counts were performed either on a hemocytometer or Cellometer.

Gene expression analysis

RNA was prepared using standard methods and hybridized to the Affymetrix GeneChip Zebrafish Genome Array. Genes that were significantly different between skin and melanoma were filtered by fold change (5

) and p-value <0.05 using the Benjamini–Hochberg false discovery rate. The up- and downregulated gene lists for each tumor were then compared with the human melanoma samples using GSEA. Enrichment of the zebrafish melanomas in the human GSE3189 dataset (7 normal skin and 45 melanomas) was identified using permutation analysis and a false discovery rate of q

0.05. After normalization and statistical analysis, genes significantly altered in NRAS1 or NRAS2 tumors were used to query the Ingenuity Pathway Analysis Package (www.ingenuity.com/). Gene lists were stratified by fold change and p-value when compared with the control skin samples, and canonical and signaling pathways found to show a significant enrichment (p

0.05) were identified.

Results

NRAS^Q61K ablates normal pigmentation

To study the effect of oncogenic NRAS in vivo, we generated transgenic strains of zebrafish expressing an NRAS^Q61K fusion protein under the control of the melanocyte-restricted zebrafish mitfa promoter (Fig. 1A). To more easily observe the expression of the transgene, we engineered an EGFP-NRAS^Q61K fusion protein.²¹ One-cell stage embryos were injected with the plasmid DNA and raised to adulthood. Although mitfa is expressed beginning at 16

h postfertilization during zebrafish development, no difference in embryonic pigment pattern was observed, a result consistent with our laboratory's previous work with BRAF^V600E. Additionally, no fluorescence was observed in injected embryos. This lack of fluorescence is likely due to photoblocking effect of melanin or a delay in EGFP folding caused by the attached NRAS protein sequence.

FIG. 1.

NRAS^Q61K causes hyperpigmentation in zebrafish. (A) Schematic representation of the transgene. (B) Animals injected with the NRAS construct develop nevi. (C) Hyperpigmentation in the NRAS1 and NRAS2 transgenic animals disrupts the normal pigment pattern. (more ...)

Injected embryos were allowed to grow to adulthood and screened for alterations in pigment pattern. At 3–4 weeks of age, ~21% of the injected fish began developing spots of hyperproliferative melanocytes that persist until adulthood (Fig. 1B). Although animals with these spots never developed melanoma, the pigmented lesions in these fish continue to slowly spread throughout the lifetime of the animals. Consistent with the photoblocking effects of melanin, nevi were not fluorescent. Mosaic transgenic fish were in-crossed to identify stable transgenic strains. The progeny of these mosaic fish was screened by PCR with primers designed against the EGFP:NRAS^Q61K transgene. Two stable transgenic lines, NRAS1 and NRAS2, were identified and represent a rate of ~2% germline transgenesis (Fig. 1C). Although both strains are similar, in that they overexpress oncogenic human NRAS, they are unique.

Both NRAS1 and NRAS2 transgenic fish display a hyperpigmented phenotype as compared with the wild-type fish that is observable at 3–4 weeks of age, a critical time point for the development of adult pigment patterning.²² The hyperpigmentation phenotypes of both strains are completely penetrant and are inherited in a dominant manner, consisting of single integration sites and Mendelian inheritance. Normal zebrafish stripe patterning is achieved by the combined effects of the epidermal, scale-associated, melanocytes along the dorsum of the animal along with the dermal melanocytes that form the lateral stripes.²² NRAS1 fish are severely hyperpigmented with the complete ablation of the normal striped pigmentation pattern. NRAS2 fish, however, have a less severe pigmentation defect and maintain the ventral-most lateral stripes. Both of these transgenic lines have a more severe pigment phenotype than Tg(mitfa:BRAF^V600E) fish.

To assess the transgene copy number in our fish, we performed quantitative RT-PCR. Our data indicated that NRAS1 fish have ~12 copies per genome, whereas NRAS2 fish have over 60 (Fig. 1D). Consistent with these results, NRAS2 fish also have significantly higher levels of the transgenic NRAS transcript than NRAS1 animals (Fig. 1E). Despite the presence of oncogenic RAS, NRAS1 and NRAS2 fish do not rapidly develop melanomas. They do, however, develop low-grade neoplasias initially observable at 1 year of age that approaches 50% penetrance by year 2 (25/50) (Fig. 1F). These tumors arise internally, dorsal to anal fin and anterior to the caudal peduncle. Although they do express the NRAS transgene, as observed by GFP fluorescence, these tumors do not spread to other parts of the animal.

NRAS^Q61K cooperates with p53 loss to generate melanoma

Although the TP53 locus is rarely mutated or lost in human melanoma, disruptions in other members of the p53 pathway are common. We therefore sought to determine if a disruption in p53 would accelerate the tumor onset or alter the tumor phenotype in our transgenic lines. We crossed NRAS1 and NRAS2 fish into the zebrafish zTP53^M214K (p53^−/−) mutant and monitored the animals for tumors. As expected, loss of one or both copies of zTP53 in both lines accelerated tumor onset and resulted in a more advanced melanoma phenotype (Fig. 2A). Tumors from both lines were fluorescent, indicating expression of the EGFP:NRAS^Q61K fusion (Fig. 2B). Both transgenic lines developed tumors with similar kinetics, with p53^−/− fish developing tumors most rapidly (Fig. 2C, D). Although p53-null tumors from both lines arose in locations that normally have melanocytes, the locations of tumors locations varied greatly between the strains (Fig. 2E, F). Whereas NRAS1;p53^−/− fish acquired tumors mostly on the fins, NRAS2;p53^−/− tumors developed more with more frequency along the body. Nearly 58% of NRAS1;p53^−/− tumors developed on the caudal fin, whereas 48% of NRAS2;p53^−/− tumors developed laterally. Tumors that arose in p53 heterozygous animals were harvested along with normal fin tissue from the same animal for DNA purification. The p53 locus was sequenced, and the loss of heterozygosity of the wild-type allele was found in 10/10 NRAS1 tumors and 4/7 NRAS2 tumors (Fig. 2G, H), indicating the importance of p53 loss in tumor progression.

FIG. 2.

NRAS^Q61K cooperates with p53 loss in melanoma formation. (A) Tumor-bearing fish were identified by visual inspection and checked for fluorescence. The box around the tumor is shown in (B) under fluorescent light. (C, D) Tumor-free survival curves are (more ...)

Tumor-bearing animals were sacrificed, and histological examination was performed (Fig. 3). Hematoxylin and eosin–stained sections revealed melanocytic tumors with a high degree of nuclear pleomorphism and the presence of melanin-bearing cells (Fig. 3A, B). As melanomas are a rare, but observed, lesion in zP53^M214K fish, we sought to confirm that our transgene was promoting melanoma in our fish. Immunohistochemistry using an anti-GFP antibody confirmed the ubiquitous presence of the EGFP:NRAS^Q61K fusion protein in these melanomas (Fig. 3C, D). To confirm a diagnosis of melanoma, we probed for mitfa mRNA, a gene that has been used clinically for diagnosis and has been shown to be uniquely required for survival of the melanocyte lineage. Consistent with human melanoma, tumors generated in NRAS1 and NRAS2 fish express mitfa (Fig. 3E, F).

FIG. 3.

Histological examination of NRAS-driven tumors in zebrafish. (A, B) Hematoxylin and eosin (H&E) staining of melanomas reveals lesions with sporadic pigmentation and high levels of nuclear pleomorphism. (C, D) Standard immunohistochemistry against (more ...)

To test the proliferative potential and malignant nature of the tumors, we performed transplantation experiments (Fig. 3G). GFP-positive cells from NRAS1:p53^−/− and NRAS2:p53^−/− tumors were isolated and transplanted intramuscularly into irradiated recipients. Sorted tumors were transplanted in limiting dilutions to assess the relative transplantibility of each tumor. Although tumors from both lines are capable of robust engraftment, RAS1 tumors require far fewer cells to engraft. Whereas as few as 1000 NRAS1 tumors are capable of engraftment, at least 50,000 NRAS2 cells are required for transplantation.

GSEA identifies molecular similarities between zebrafish and human melanoma

To determine if the gene expression profile of our RAS-driven zebrafish melanomas is molecularly similar to the human disease, we performed GSEA. GSEA is a computational method of determining whether a predetermined gene set is enriched in a biological state, when compared with another biological state.²³ One advantage of this method is that it allows for the comparison of gene lists that are derived from different microarray platforms and, importantly, different species. This method has been used successfully to show that zebrafish tumors of the liver and muscle share conserved molecular signatures with their counterpart human diseases.^24–26 We sought to perform a similar analysis to determine if the pathways activated in zebrafish melanoma are similarly regulated in the human disease.

RNA from three normal zebrafish skin samples and six zebrafish melanomas, from three NRAS1;p53^−/− tumors and three NRAS2;p53^−/− tumors, was hybridized to the Affymetrix zebrafish genome array. Lists of the up- and downregulated genes were generated by comparing the microarray data from normal zebrafish skin to zebrafish melanoma (Supplemental Table S1, available online at www.liebertonline.com). The gene lists resulting from this comparison were limited to genes that showed greater than fivefold change between skin and tumor with p-values <0.05. Similar manipulations were also done with the human GSE3189 microarray dataset, which includes data from 7 normal skin samples and 45 melanomas.²⁷ Because activating mutations in NRAS are exceptionally rare in human nevi, we did not include the GSE3189 data on nevi in our analysis. Instead, we focused on the comparison to normal human skin and human melanoma.

Using GSEA, we determined whether the up- and downregulated zebrafish gene lists (from both NRAS1;p53^−/− and NRAS2;p53^−/−) were statistically associated with human melanoma as compared with the human skin. Interestingly, both zebrafish upregulated gene sets were significantly associated with the human melanoma dataset. To illustrate the similarity, the NRAS1;p53^−/− upregulated genes were used to sort microarray data from human melanoma and normal human skin (Fig. 4A). The genes that are highly expressed in zebrafish melanoma are similarly overexpressed in human disease, but not normal human skin. This indicates that the list of genes that are upregulated in zebrafish melanomas is similar to the list of genes that are upregulated in human melanoma. The NRAS1;p53^−/− GSEA gene list when compared with the human gene list shows a statistically significant false discovery rate q-value of 0.023. The enrichment plot for downregulated NRAS1;p53^−/− is also shown (Fig. 4D). The downregulated gene list was not significantly associated with human melanoma (q

0.26). The enrichment plot for the upregulated NRAS2;p53^−/− gene list is also shown (Fig. 4C). Like the NRAS1;p53^−/− gene list, only the upregulated NRAS2;p53^−/− gene list was significantly associated with human melanoma (q

0.004). Collectively, these data indicate that the pathways activated in melanoma are conserved between zebrafish and human melanoma.

FIG. 4.

Gene set enrichment analysis identifies upregulated gene lists that are conserved between zebrafish and human melanoma. The zebrafish gene set enrichment analysis gene list was used to parse human skin and melanoma microarrays. The parsed list is shown (more ...)

Although the NRAS1 and NRAS2 tumors show considerable transcriptional overlap, distinct pathway activation could be discerned between the tumor types. The NRAS1 tumors show significant enrichment in components of the Wnt/beta-catenin pathway as well as genes involved in melanocyte development and pigmentation (Supplemental Fig. S1, available online at www.liebertonline.com). This is not entirely unexpected, given the phenotype of the tumors. In contrast, the NRAS2 tumors show a significant number of genes associated with cell–cell communication, calcium signaling, and MAPK signaling. These data may suggest that the differences in gene dosage may have significant effects on melanoma biology not only at the phenotypic level but also at the transcriptional level.

Discussion

The pigment phenotype found in the NRAS1 and NRAS2 transgenic fish is striking yet similar to the observed hyperpigmentation in BRAF^Q61K transgenic fish. Activation of the MAPK pathway in zebrafish melanocytes clearly has clear effects on patterning. The difference in pigmentation and DNA content between the two transgenic lines is intriguing, especially because the less severe fish phenotypically (RAS2) has higher levels of RAS expression. This result agrees with the previous observations that extremely high levels of RAS expression can have different cellular effects than lower levels of RAS expression. Specifically, high-intensity RAS signaling has been shown to activate senescence pathways in addition to MAPK signaling.²⁸ In these studies, it was found that high-intensity, but not low-intensity, RAS signaling leads to senescence by activating MEKK3/6-p38, which leads to an increase in Ink4a transcription. It is possible that the higher levels of transgene expression in RAS2 fish prevent more severe cellular and pigmentation phenotypes by contributing to premature cell-cycle arrest. Although we were unable to detect apoptosis or senescence in adult scale-associated melanocytes, it is possible that with improved assays we would be able to see this phenotype in 3–4 week juveniles, the age when hyperpigmentation becomes apparent.

Although NRAS1 and NRAS2 fish express oncogenic NRAS, they do not readily develop cancer. This result indicates that in fish, analogous to the mammalian systems, the presence of a single oncogenic stimulus is insufficient for tumorigenesis. Indeed, the observation that p53-deficient transgenic animals do not develop cancer until at least 10 weeks of age further confirms a multihit tumor model and the need to accumulate additional genomic alterations. It is interesting that the rate of tumor onset in p53-matched fish, either RAS1;p53^−/− versus RAS2;p53^−/− or RAS1;p53^+/− versus RAS2;p53^+/−, is similar. This suggest that there may be a threshold effect for RAS signaling in zebrafish melanocytes, above which excess NRAS does not impart an additional oncogenic stimulus. Instead, in the context of oncogenic NRAS coupled with p53 loss, the rate of tumor onset is dependent on the accumulation of spontaneous mutations. The multihit nature of this tumor model further underscores its similarity with human disease.

In the p53 heterozygous background, tumor progression requires the loss of heterozygosity of the remaining wild-type p53 allele. The fact that wild-type p53 is selectively lost in these fish confirms the required role in zebrafish for p53 in genome maintenance and cell-cycle regulation. The observation that p53 is lost in 100% of NRAS1;p53^+/−-derived tumors but only 57% of NRAS2;p53^+/−-derived tumors is not a significant difference (p

0.3) and likely reflects technical challenges associated with the difference in tumor locations instead of different mechanisms of tumor progression. NRAS1 tumors were collected from fins and were therefore easy to dissect away from normal tissue. NRAS2 tumors, however, arose laterally, and clean dissections were often not possible.

In addition to the histological similarities, whole-genome expression arrays indicate that NRAS1;p53^−/− and NRAS2;p53^−/− tumors have high degrees of molecular similarity to human disease. Not only does this indicate that the transcriptional pathways activated in human melanoma are conserved in zebrafish but also it validates the utility of zebrafish melanoma as a model for studying human disease. Although both NRAS models develop tumors highly similar to human melanomas, they are unique from each other. Although the similarities between the two models may provide insights into some of the more common features of human disease, exploring the differences may be equally useful in studying relevant subclasses of human disease.

It is interesting to note that only upregulated genes are conserved between zebrafish and human melanoma. Although there are many possible causes, a major contributor is likely the fact that normal skin was used for the initial comparisons to melanoma. Ideally, purified melanocytes would have been used as the normal tissue control. However, because zebrafish melanocytes cannot be purified, this was not possible. With significantly fewer melanocytes in normal skin than melanoma, even modest increases in expression would appear amplified in melanoma. Additionally, there is effectively no upper limit to overexpression. In an attempt to control for these issues, zebrafish melanoma arrays were compared with the mean observed expression of the chip. The most highly expressed genes on the chip were the same genes identified in our original analysis.

In contrast, detection of downregulated genes may require purified melanocytes as a control. A normally expressed gene in a relatively small number of normal melanocytes may simply resemble, on an array, the expression profile for that gene when downregulated in a melanoma sample. Plus, there is a lower limit on downregulation. If a probe set registers zero when data are collected, it cannot be evaluated and is automatically excluded from analysis. This phenomenon that only upregulated genes are conserved between human and zebrafish cancer has been observed previously in other zebrafish cancer models and may reflect a limitation of the system or the assay.²⁵

In summary, we have described two new zebrafish models of NRAS-driven human melanoma. We found that when expressed in the melanocyte compartment, human oncogenic NRAS^Q61K causes hyperpigmentation in zebrafish and cooperates with p53 loss to generate melanomas. The phenotypic similarities between zebrafish melanoma and human disease underscore both the importance and the utility of further zebrafish research. Further, the similarities in gene expression between human and zebrafish disease may provide clues as to what aspects of human melanoma will be most amenable to dissection through zebrafish models.

Supplementary Material

Supplemental data

Click here to view.^{(169K, pdf)}

Acknowledgments

We would like to thank members of the Zon laboratory, especially Craig Ceol, for helpful discussions. R.M.W. is supported by an American Society for Clinical Oncology Young Investigator Award and an Aid for Cancer Research Fellowship. Melanoma research by L.I.Z. is funded by National Institutes of Health Grant R01-DK53298-08.

Disclosure Statement

No competing financial interests exist.

References

1. ACS Detailed Guide: Skin Cancer—Melanoma. 2008. www.cancer.org. www.cancer.org

2. Chin L. Garraway LA. Fisher DE. Malignant melanoma: genetics and therapeutics in the genomic era. Genes Dev. 2006;20:2149–2182. [PubMed]

3. Jiveskog S. Ragnarsson-Olding B. Platz A. Ringborg U. N-ras mutations are common in melanomas from sun-exposed skin of humans but rare in mucosal membranes or unexposed skin. J Invest Dermatol. 1998;111:757–761. [PubMed]

4. Omholt K. Platz A. Kanter L. Ringborg U. Hansson J. NRAS and BRAF mutations arise early during melanoma pathogenesis and are preserved throughout tumor progression. Clin Cancer Res. 2003;9:6483–6488. [PubMed]

5. Tsao H. Goel V. Wu H. Yang G. Haluska FG. Genetic interaction between NRAS and BRAF mutations and PTEN/MMAC1 inactivation in melanoma. J Invest Dermatol. 2004;122:337–341. [PMC free article] [PubMed]

6. Demunter A. Stas M. Degreef H. De Wolf-Peeters C. van den Oord JJ. Analysis of N- and K-ras mutations in the distinctive tumor progression phases of melanoma. J Invest Dermatol. 2001;117:1483–1489. [PubMed]

7. Jafari M. Papp T. Kirchner S. Diener U. Henschler D. Burg G. Schiffmann D. Analysis of ras mutations in human melanocytic lesions: activation of the ras gene seems to be associated with the nodular type of human malignant melanoma. J Cancer Res Clin Oncol. 1995;121:23–30. [PubMed]

8. Bos JL. ras oncogenes in human cancer: a review. Cancer Res. 1989;49:4682–4689. [PubMed]

9. Chudnovsky Y. Adams AE. Robbins PB. Lin Q. Khavari PA. Use of human tissue to assess the oncogenic activity of melanoma-associated mutations. Nat Genet. 2005;37:745–749. [PMC free article] [PubMed]

10. Wan PT. Garnett MJ. Roe SM. Lee S. Niculescu-Duvaz D. Good VM, et al. Mechanism of activation of the RAF-ERK signaling pathway by oncogenic mutations of B-RAF. Cell. 2004;116:855–867. [PubMed]

11. Michaloglou C. Vredeveld LC. Soengas MS. Denoyelle C. Kuilman T. van der Horst CM, et al. BRAFE600-associated senescence-like cell cycle arrest of human naevi. Nature. 2005;436:720–724. [PubMed]

12. Papp T. Pemsel H. Zimmermann R. Bastrop R. Weiss DG. Schiffmann D. Mutational analysis of the N-ras, p53, p16INK4a, CDK4, and MC1R genes in human congenital melanocytic naevi. J Med Genet. 1999;36:610–614. [PMC free article] [PubMed]

13. Albino AP. Nanus DM. Mentle IR. Cordon-Cardo C. McNutt NS. Bressler J. Andreeff M. Analysis of ras oncogenes in malignant melanoma and precursor lesions: correlation of point mutations with differentiation phenotype. Oncogene. 1989;4:1363–1374. [PubMed]

14. Patton EE. Widlund HR. Kutok JL. Kopani KR. Amatruda JF. Murphey RD, et al. BRAF mutations are sufficient to promote nevi formation and cooperate with p53 in the genesis of melanoma. Curr Biol. 2005;15:249–254. [PubMed]

15. Chin L. Pomerantz J. Polsky D. Jacobson M. Cohen C. Cordon-Cardo C, et al. Cooperative effects of INK4a and ras in melanoma susceptibility in vivo. Genes Dev. 1997;11:2822–2834. [PubMed]

16. Ha L. Ichikawa T. Anver M. Dickins R. Lowe S. Sharpless NE, et al. ARF functions as a melanoma tumor suppressor by inducing p53-independent senescence. Proc Natl Acad Sci USA. 2007;104:10968–10973. [PubMed]

17. Hussussian CJ. Struewing JP. Goldstein AM. Higgins PA. Ally DS. Sheahan MD, et al. Germline p16 mutations in familial melanoma. Nat Genet. 1994;8:15–21. [PubMed]

18. Ackermann J. Frutschi M. Kaloulis K. McKee T. Trumpp A. Beermann F. Metastasizing melanoma formation caused by expression of activated N-RasQ61K on an INK4a-deficient background. Cancer Res. 2005;65:4005–4011. [PubMed]

19. Ballester M. Castello A. Ibanez E. Sanchez A. Folch JM. Real-time quantitative PCR-based system for determining transgene copy number in transgenic animals. Biotechniques. 2004;37:610–613. [PubMed]

20. Lister JA. Robertson CP. Lepage T. Johnson SL. Raible DW. nacre encodes a zebrafish microphthalmia-related protein that regulates neural-crest-derived pigment cell fate. Development. 1999;126:3757–3767. [PubMed]

21. Choy E. Chiu VK. Silletti J. Feoktistov M. Morimoto T. Michaelson D, et al. Endomembrane trafficking of ras: the CAAX motif targets proteins to the ER and Golgi. Cell. 1999;98:69–80. [PubMed]

22. Rawls JF. Mellgren EM. Johnson SL. How the zebrafish gets its stripes. Dev Biol. 2001;240:301–314. [PubMed]

23. Subramanian A. Tamayo P. Mootha VK. Mukherjee S. Ebert BL. Gillette MA, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci USA. 2005;102:15545–15550. [PubMed]

24. Lam SH. Wu YL. Vega VB. Miller LD. Spitsbergen J. Tong Y, et al. Conservation of gene expression signatures between zebrafish and human liver tumors and tumor progression. Nat Biotechnol. 2006;24:73–75. [PubMed]

25. Langenau DM. Keefe MD. Storer NY. Guyon JR. Kutok JL. Le X, et al. Effects of RAS on the genesis of embryonal rhabdomyosarcoma. Genes Dev. 2007;21:1382–1395. [PubMed]

26. Shepard JL. Amatruda JF. Stern HM. Subramanian A. Finkelstein D. Ziai J, et al. A zebrafish bmyb mutation causes genome instability and increased cancer susceptibility. Proc Natl Acad Sci USA. 2005;102:13194–13199. [PubMed]

27. Talantov D. Mazumder A. Yu JX. Briggs T. Jiang Y. Backus J, et al. Novel genes associated with malignant melanoma but not benign melanocytic lesions. Clin Cancer Res. 2005;11:7234–7242. [PubMed]

28. Deng Q. Liao R. Wu BL. Sun P. High intensity ras signaling induces premature senescence by activating p38 pathway in primary human fibroblasts. J Biol Chem. 2004;279:1050–1059. [PubMed]

Articles from Zebrafish are provided here courtesy of
Mary Ann Liebert, Inc.