Metastasis is inevitably a collaboration between genetic/epigenetic alterations present in the tumor cells along with the microenvironment that supports metastatic outgrowth. The zebrafish is uniquely suited to perform high-throughput, high-content screens for factors in either compartment that modulate metastatic frequency. The information from these screens provides basic insights into the process, and can also act as a platform for novel therapeutic discovery to prevent or ameliorate metastatic progression.
Cancer modeling in the zebrafish
The zebrafish is highly amenable to transgenic and carcinogenic tumor formation, and can be used to study a wide variety of cancer mechanisms (Figure 1). We have developed a zebrafish model of melanoma in which the mutant human BRAFV600E allele is expressed under the melanocyte specific mitf promoter. When crossed with p53 mutant fish, the resultant BRAF;p53 animals all develop melanoma within four to 12 months (Figure 2). These tumors strongly resemble human melanoma at both the histological and gene expression levels, and serve as a platform for unbiased genetic and small molecule screening approaches. In addition to transgenic models of melanoma, fish are susceptible to transposon and retroviral induction of tumors in a wide variety of organs such as the pancreas, blood, and liver.
Studying metastasis in transparent zebrafish
Metastasis is a multistep process, and imaging and isolation of cells at each step is required to understand the factors contributing to each stage. To enhance this ability, we developed a zebrafish strain named casper (Figure 3), which maintains its transparency throughout adulthood. The casper model provides exquisite single-cell analysis of metastatic dissemination at each step in the cascade. We have utilized this model to identify genes required for dissemination of T cell lymphoma and melanoma cells, and to determine how these cells interact with the host stroma and vasculature.
We have developed quantitative systems for analyzing metastasis in the zebrafish, taking advantage of the optical clarity of the casper strain. By transplanting small numbers of cells into either larval or adult zebrafish, we are able to image metastatic progression over time and space (Figure 4). We have developed tools that allow us to quantify the degree of metastatic spread in different groups of fish, opening the way to performing high-throughput, high-content metastasis screens in vivo.
Developmental lineage factors in melanoma
We have used the transgenic melanoma model in a chemical screen to find modulators of melanoma growth. Using gene expression analysis, we identified an in vivo signature of melanoma initiation, which was highly enriched for markers of the embryonic neural crest such as sox10 and mitf.. This is the embryonic lineage from which all pigmented melanocytes are derived. By testing a library of small molecules in >30,000 zebrafish, we identified leflunomide, an inhibitor of the metabolic enzyme DHODH, as a strong in vivo modifier of this neural crest transcription program. This molecule has strong effects on melanoma growth, indicating that we can target not only mutations in cancer cells, but also the cellular “milieu” in which mutations such as BRAFV600E exist. This screen was the first to demonstrate the possibility of therapeutically targeting lineage factors in melanoma. We are now investigating how these developmental lineage factors are utilized during each step in the metastatic cascade, using both transgenic and transplantation based approaches. Because several of these neural crest programs can be targeted by small molecules, these pathways represent attractive therapeutic targets in halting metastatic progression.
Microenvironmental factors in melanoma
Because cell-intrinsic changes are only a part of metastatic progression, we are now investigating which cells in the microenvironment can promote or retard metastatic growth. Melanoma cells are surrounded by numerous microenvironmental cell types such as keratinocytes and adipocytes, and targeting the cross-talk between these cell types is an important factor in regulating metastatic growth. Using CRISPR and cell ablation techniques, we are screening for factors that regulate this interplay during each stage of metastasis.
We are using the zebrafish to evaluate novel mechanisms of genetic evolution in melanoma. Towards this end, we have now undertaken several large-scale efforts aimed at defining the genome of zebrafish melanoma (Figure 6). This investigation includes exome sequencing as well as whole-genome and RNA-seq analysis of matched DNA from primary and metastatic tumors, and normal germline DNA. Our research has revealed that zebrafish tumors undergo large-scale genomic alterations similar to those occurring in human tumors, with a particular conservation of copy number alterations across the two species. The development of these genomic platforms allows for an in-depth understanding of how tumor genomes evolve over time, and how these genomes evolve during metastatic progression.