(91f) A CRISPR-Cas9 Screen for Systematic Discovery of Genes and Pathways Controlling Human Melanogenesis

Melanin, an inert biopolymer, is responsible for skin, hair and eye pigmentation. Consequently, genetic variants affecting melanogenesis underlie diversification of human species into distinctly pigmented subpopulations. Melanin production also contributes to the natural homeostasis of tissues that produce it, and dysregulation of these pathways can lead to slew of human pathological conditions affecting the skin (e.g. vitiligo and melasma), eyes (age related macular degeneration) and ear (sensorineural deafness) ^1,2. Melanin is typically produced by melanocytes, which originate from the neural crest and reside in the epidermis, where they transport melanin filled vesicles (a.k.a. melanosomes) to the surrounding keratinocytes, resulting in skin pigmentation. Importantly, the number of melanocytes and their anatomical location do not vary among humans, only varying melanogenesis determines the human pigmentation diversity^2,3. The core biochemical machinery that regulates melanin synthesis has been known for a long time and further key insights have come from mapping genes involved in hypo- and hyper-pigmentation diseases and from genome-wide association studies (GWAS) of normal-range skin and hair color variation in human populations³. However, no systematic discovery of genes involved in human melanogenesis has been performed to date, and current candidate GWAS variants can explain only a fraction of skin color variation in humans, suggesting that many loci controlling pigmentation remain undiscovered. In this study, we performed a genome-wide CRISPR-Cas9 screen to systematically discover genes controlling human melanogenesis.

To this end, we first demonstrate a tight correlation between melanin content of pigment cells and their light scattering properties as measured as side scatter parameter (SSC) by flow cytometry. To establish a link between melanin content and SSC, we modeled human melanogenesis in vitro using pluripotent stem cells in a step-wise manner where unpigmented MC precursors acquired melanin as they matured into pigmented MC. By measuring melanin content of the pigment cells and SSC during this maturation process, we determined a strong correlation between SSC and melanin (R² = 0.82). We next exploited this relationship to directly survey genes involved in melanogenesis through a genome-wide CRISPR screen, with an advantage of not being confounded by the indirect factors influencing pigmentation by impacting melanocyte development or survival. To this end, we generated a clonal MNT1 melanoma cell line inducibly expressing spCas9 nuclease (Cas9-MNT1 cell line) upon doxycycline treatment, and infected it with a genome-wide lentiviral sgRNA library (with 10 targeting-guides per gene) ⁴, such that each cell expressed a single sgRNA. In addition, the library also included negative control sgRNAs, such as non-targeting guides (i.e. no binding sites in the genome) and safe-targeting control guides targeting genomic locations with no annotated function⁴. Following puromycin selection to kill noninfected cells, expression of Cas9 was induced to carry out gene-editing, and after two weeks, top 10% low and high SSC cells were FACS sorted. As expected, low SSC fraction was enriched for hypo-pigmented cells. Genomic DNA was subsequently isolated and the frequencies of sgRNAs in both populations (low and high SSC) were measured by deep sequencing and analyzed using Cas9 high-Throughput maximum Likelihood Estimator (CasTLE) algorithm⁵. Analysis of sgRNA insertion frequencies in two independent biological replicates (i.e. independent sgRNA infections and FACS sorting post Cas9 induction) revealed good concordance (R² = 0.59) between the replicates and our screen identified 169 genes whose deletion results in diminished melanin content (e.g. melanin-promoting genes), 23 of them corresponding to previously known regulators of pigmentation (including ‘famous’ pigmentation genes such as TYR, OCA2, SLC45A2, SLC24A5, and LYST). However, most of the screen hits represent novel associations with pigmentation and are involved in diverse cellular functions, converging on vesicle organization, endosomal transport and transcriptional/posttranscriptional gene regulation. We further validated top hits of our genome-wide screen, by deleting 10 novel candidate genes individually using CRISPR/Cas9 editing. By serial transmission electron microscopy (TEM) studies, we further demonstrated that loss of gene function affected melanosome composition/biogenesis thus resulting in reduced melanin phenotype of pigment cells. Next, we collected lists of genes implicated by various published skin pigmentation GWAS and found that relative to all other genes assayed, our CRISPR screen hits were strongly enriched for pigmentation genes across the different GWAS datasets further suggesting putative role of novel genes in human skin color variation. Finally, we isolated melanocytes from skin of diversely pigmented newborn males, and through transcriptome (RNA-seq) analysis show that the majority of genes discovered in our screen are upregulated in dark skin melanocytes, in agreement with their melanin-promoting function and potential contribution to skin color variation in humans.

By combining melanin’s structure and functional properties with CRISPR tools, we identified novel human pigmentation genes that regulate melanogenesis and influence normal range skin color variation in humans and this useful resource would catalyze more studies on human pathophysiological pigmentation

1 Yamaguchi, Y., Brenner, M. & Hearing, V. J. The regulation of skin pigmentation. The Journal of biological chemistry 282, 27557-27561, doi:10.1074/jbc.R700026200 (2007).

2 Sturm, R. A. Molecular genetics of human pigmentation diversity. Hum Mol Genet 18, R9-17, doi:10.1093/hmg/ddp003 (2009).

3 Pavan, W. J. & Sturm, R. A. The Genetics of Human Skin and Hair Pigmentation. Annu Rev Genomics Hum Genet 20, 41-72, doi:10.1146/annurev-genom-083118-015230 (2019).

4 Morgens, D. W. et al. Genome-scale measurement of off-target activity using Cas9 toxicity in high-throughput screens. Nature communications 8, 15178, doi:10.1038/ncomms15178 (2017).

5 Morgens, D. W., Deans, R. M., Li, A. & Bassik, M. C. Systematic comparison of CRISPR/Cas9 and RNAi screens for essential genes. Nat Biotechnol 34, 634-636, doi:10.1038/nbt.3567 (2016).