The cumbersome and time-consuming process of generating new mouse strains and multiallelic experimental animals often hinders the use of genetically engineered mouse models (GEMM) in cancer research. Here, we describe the development and validation of an embryonic stem cell (ESC)-GEMM platform for rapid modeling of melanoma in mice. The platform incorporates 12 clinically relevant genotypes composed of combinations of four driver alleles (LSL-Braf V600E , LSL-Nras Q61R , Pten Flox , and Cdkn2a Flox ) and regulatory alleles to spatiotemporally control the perturbation of genes of interest. The ESCs produce high-contribution chimeras, which recapitulate the melanoma phenotypes of conventionally bred mice. Using the ESC-GEMM platform to modulate Pten expression in melanocytes in vivo, we highlighted the utility and advantages of gene depletion by CRISPR-Cas9, RNAi, or conditional knockout for melanoma modeling. Moreover, complemen-tary genetic methods demonstrated the impact of Pten restoration on the prevention and maintenance of Pten-deficient melanomas. Finally, we showed that chimera-derived melanoma cell lines retain regulatory allele competency and are a powerful resource to complement ESC-GEMM chimera experiments in vitro and in syngeneic grafts in vivo. Thus, when combined with sophisticated genetic tools, the ESC-GEMM platform enables rapid, high-throughput, and versatile studies aimed at addressing outstanding questions in melanoma biology.Significance: This study presents a high-throughput and versatile ES cell-based mouse modeling platform that can be combined with state-of-the-art genetic tools to address unanswered questions in melanoma in vivo.See related commentary by Thorkelsson et al., p. 655

Here we present an innovative computational-based drug discovery strategy, coupled with machine-based learning and functional assessment, for the rational design of novel small molecule inhibitors of the lipogenic enzyme stearoyl-CoA desaturase 1 (SCD1). Our methods resulted in the discovery of several unique molecules, of which our lead compound SSI-4 demonstrates potent anti-tumor activity, with an excellent pharmacokinetic and toxicology profile. We improve upon key characteristics, including chemoinformatics and absorption/distribution/metabolism/excretion (ADME) toxicity, while driving the IC50 to 0.6 nM in some instances. This approach to drug design can be executed in smaller research settings, applied to a wealth of other targets, and paves a path forward for bringing small-batch based drug programs into the Clinic.

The cumbersome and time-consuming process of generating new mouse strains and multi-allelic experimental animals often hinders the use of genetically engineered mouse models (GEMM) in cancer research. Here, we describe the development and validation of an embryonic stem cell (ESC)-GEMM platform for rapid modeling of melanoma in mice. Our platform incorporates twelve clinically relevant genotypes composed of combinations of four driver alleles (LSL-BrafV600E, LSL-NrasQ61R, PtenFlox, Cdkn2aFlox) and regulatory alleles to spatiotemporally control the perturbation of genes-of-interest. Our ESCs produce high contribution chimeras, which recapitulate the melanoma phenotypes of conventionally bred mice. Using our ESC-GEMM platform to modulate Pten expression in melanocytes in vivo, we highlight the utility and advantages of gene depletion by CRISPR-Cas9, RNAi, or conditional knockout for melanoma modeling. Moreover, we use complementary genetic methods to demonstrate the impact of Pten restoration on the prevention and maintenance of Pten-deficient melanomas. Finally, we show that chimera-derived melanoma cell lines retain regulatory allele competency and are a powerful resource to complement ESC-GEMM chimera experiments in vitro and in syngeneic grafts in vivo. Thus, when combined with sophisticated genetic tools, our ESC-GEMM platform enables rapid, high-throughput, and versatile studies aimed at addressing outstanding questions in melanoma biology.