FRA1 drives melanoma metastasis through an actionable transcriptional network

Abstract

Transcriptional dysregulation has emerged as a critical driver of melanoma progression, yet the molecular mechanisms governing this process and their potential as therapeutic targets remain inadequately characterized. Here, we identify FRA1 as a potent and actionable driver of melanoma metastasis. FRA1 enhanced both the initial seeding and subsequent outgrowth of metastatic lesions. Comprehensive multi-omics integration revealed transcriptional target genes of FRA1, with AXL, CDK6, and FSCN1 exhibiting increased expression in melanoma metastasis and a significant correlation with poor patient outcomes. Silencing AXL, CDK6, or FSCN1 abrogated FRA1-mediated invasion in vitro and reduced metastatic colonization. Furthermore, pharmacological inhibition of CDK6 and FSCN1, and to a lesser extent AXL, suppressed melanoma metastasis and prolonged overall survival. The expression of FRA1 and its target genes correlates with shortened survival across multiple cancer types, highlighting the broader clinical relevance of this pathway. This study unveils an actionable FRA1-mediated transcriptional network that drives cancer progression and metastasis, offering potential avenues for therapeutic interventions.

Fig. 1. FRA1 drives melanoma metastasis.

A Representative images of IHC showing the expression of FRA1 in a melanoma tissue microarray containing primary melanoma and metastatic melanoma specimens. Bar plot showing frequency of FRA1 expression patterns (Strong positive: FRA1 is expressed in >30% cells; Weak positive: FRA1 is expressed in 5%–30% cells; Negative: FRA1 is expressed in <5% cells) in primary skin melanomas and lymph node metastatic melanomas. B 1205Lu melanoma cells overexpressing FRA1 were subcutaneously injected into NSG mice (n = 10). Tumor volumes were measured every 7 days. C, D 1205Lu melanoma cells with inducible CRISPR interference (CRISPRi) targeting FRA1 were subcutaneously injected into NSG mice (n = 6). Mice were fed chow containing 200 mg/kg Dox to induce CRISPRi. Tumor volumes were measured every 15 days (C). Spontaneous lung metastasis burden of 1205Lu cells by H&E staining (D). E–G Luciferase tagged 1205Lu cells with inducible CRISPRi targeting FRA1 were intravenously injected into NSG mice (n = 5) and the metastases were measured after 22 days by bioluminescent In Vivo Imaging System (IVIS) (E). Luciferase tagged 1205Lu and M10M6 cells with FRA1 overexpression were intravenously injected into NSG mice (n = 5). The metastases were measured after 18 days by IVIS (F and G). Representative bioluminescence images and quantification of the luminescence signals are shown. H GSEA analysis of TCGA-SKCM samples associates FRA1 expression with a melanoma metastasis gene signature. Data are presented as mean ± SEM and analyzed with Student’s unpaired t test, * P < 0.05, ** P < 0.01, *** P < 0.001.

Fig. 2. FRA1 enhances metastatic colonization and outgrowth.

A–C 1205Lu melanoma cells overexpressing FRA1 were intravenously injected into NSG mice (n = 5) (A). 1205Lu and A375 cells with inducible CRISPRi targeting FRA1 were intravenously injected into NSG mice (n = 5), and mice were fed chow containing 200 mg/kg Dox to induce CRISPRi (B, C). 7 days after inoculation, metastasis was detected by H&E staining. Representative images of micro-metastasis in lungs and quantification of the number and average size of micro-metastasis are shown. D–H Luciferase tagged 1205Lu and A375 cells with inducible CRISPRi targeting FRA1 were intravenously injected into NSG mice (n = 5), and 10 days after inoculation mice were switched to chow containing 200 mg/kg Dox to induce CRISPRi, and then metastasis were analyzed 2 weeks after Dox diet feeding, which is 24 days after cell inoculation (D). Representative images and quantification of the luminescence signals of 1205Lu cells are shown in E. Representative images and quantification of lung metastasis burden of 1205Lu and A375 cells by H&E staining are shown in (F, G). Representative images and quantification of liver metastasis number and burden of A375 cells are shown in H. Data are presented as mean ± SEM and analyzed with Student’s unpaired t test, * P < 0.05, ** P < 0.01, *** P < 0.001, ns not significant.

Fig. 3. Transcriptome of FRA1 in melanoma.

A–C A375 and 1205Lu cells with FRA1 silenced by siRNAs were subjected to RNA sequencing. Volcano plots show the differentially expressed genes by FRA1 in A375 cells (A) and 1205Lu cells (B). Scatter plot shows correlation of gene regulation by FRA1 in A375 and 1205Lu cells (C). D, E GSEA analysis of Hallmark gene signature showing the enrichment of genes regulated by FRA1 in A375 cells (D) and 1205Lu cells (E). F GSEA analysis showing transcriptome of FRA1 is associated with a melanoma metastasis gene signature. G Overlapped differentially expressed genes by FRA1 in A375 and 1205Lu cells.

Fig. 4. Genome-wide binding profile of FRA1 in melanoma.

A A375 and 1205Lu cells were subjected to Cut&Run sequencing by FRA1 and IgG as negative control. Heatmap showing the read density of FRA1 on the genomes of A375 cells and 1205Lu cells. Venn diagram showing the overlapped peaks. B De Novo Motif analysis by HOMER showing enriched motifs of FRA1 in melanoma cells. C Genomic annotation of overlapped FRA1 bound peaks in A375 and 1205Lu cells. D Scatter plot showing the enrichment of overlapping FRA1 peaks in A375 cells and 1205Lu cells. E, F Gene Ontology (GO) analysis of overlapping FRA1 peaks showing the enriched gene signature in GO Biological Process (E) and GO Molecular Function (F).

Fig. 5. FRA1 transcriptionally activates AXL, CDK6, and FSCN1.

A IGV genome tracks highlight FRA1 occupancy at the AXL, CDK6, and FSCN1 genomic loci. B Overall survival curves of melanoma patients with or without alteration of AXL, CDK6, or FSCN1 (including mutation, copy number alteration, and mRNA changes) in TCGA-SKCM dataset. C Western blot analysis of AXL, CDK6, and Fascin upon FRA1 overexpression or silencing in melanoma cells. D Western blot analysis of FRA1, AXL, CDK6, Fascin, and MITF in human melanocytes (H1, H2, H3, H4), human melanocytes expressing oncogenic BRAF^V600E (H1-B, H3-B), and melanoma cell lines. E Representative image of multiplex IHC showing the expression of FRA1, AXL, CDK6, and Fascin in a melanoma microarray containing primary melanoma and metastatic melanoma. F Bar plot showing frequency of Fascin positive samples in primary skin melanomas and lymph node metastatic melanomas. G Pearson correlation analysis of fluorescent intensity of FRA1 and AXL, CDK6, or Fascin. H, I Scatter plots showing the correlation between FRA1 expression and the mean expression of AXL, CDK6, FSCN1 in 1673 cancer cell lines (H, Data extracted from DepMap) and 1,210 tumor tissues (I, Data extracted from cBioPortal). Expression levels are presented as log₂(TPM + 1). J qRT-PCR showing the relative expression of AXL, CDK6, and FSCN1 in 1205Lu cells upon expression of PTEN. Data are presented as mean ± SEM and analyzed with Student’s unpaired t test, * P < 0.05, ** P < 0.01, *** P < 0.001.

Fig. 6. AXL, CDK6, and FSCN1 mediate FRA1 pro-metastatic functions.

A Proliferation as measured by relative confluence of A375 cells with FRA1 silencing and concomitant AXL, CDK6, or FSCN1 overexpression. B Quantification of cell numbers of transwell invasion assays of A375 cells with FRA1 silencing and concomitant AXL, CDK6, or FSCN1 overexpression. C Proliferation as measured by relative confluence of 1205Lu cells expressing FRA1 with AXL, CDK6, or FSCN1 silenced. D Quantification of cell numbers of transwell invasion assays of 1205Lu cells expressing FRA1 with AXL, CDK6, or FSCN1 silenced. E FRA1 overexpressing 1205Lu cells with AXL, CDK6, or FSCN1 silenced were intravenously injected into NSG mice (n = 5). 7 days after inoculation, metastasis was detected by H&E staining. Representative images of micro-metastasis in lungs and quantification of the number and average size of micro-metastasis are shown. F, G Luciferase tagged 1205Lu cells were intravenously injected into NSG mice. AXL inhibitor Bemcentinib, CDK4/6 inhibitor G1T38, Fascin inhibitor NP-G2-044, and vehicle control were intraperitoneally (I.P.) administrated to mice (n = 5) at day 14, 16, 18, and 20 after cell inoculation (F). Metastasis was measured at day 20 by IVIS, and representative images and quantification of luminescence signals are shown (G). H 1205Lu cells were intravenously injected into NSG mice. Single or combination treatment of CDK4/6 inhibitor Abemaciclib and Fascin inhibitor NP-G2-044, or vehicle control were intraperitoneally (I.P.) administrated to mice (n = 5) once per day (Q.D.) until endpoint. I Quantification of metastasis burden in individual mice measured every 5 days by IVIS. J Overall survival curves of mice treated with the indicated inhibitors or vehicle control. Data are presented as mean ± SEM and analyzed with Student’s unpaired t test, * P < 0.05, ** P < 0.01, *** P < 0.001.

Fig. 7. Evidence of oncogenic FRA1 signaling in multiple cancer types.

A Overall survival curve of patients with high or low FRA1 expression in multiple cancer types from TCGA Pan Cancer Atlas dataset. B Correlation between FRA1 expression and expression of AXL, CDK6, FSCN1 in multiple cancer types from TCGA Pan Cancer Atlas dataset. C Correlation between FRA1 expression and expression of AXL, CDK6, FSCN1 in multiple cancer types from Cancer Cell Line Encyclopedia (CCLE) (Data extracted from DepMap). (LIHC: Liver hepatocellular carcinoma. PDAC: Prostate adenocarcinoma. CODA: Colon adenocarcinoma. LUAD: Lung adenocarcinoma. LUSC: Lung squamous cell carcinoma. NSCLC: Non-small cell lung cancer. STAD: Stomach adenocarcinoma. KIRC: Kidney renal clear cell carcinoma. BLCA: Bladder urothelial carcinoma. BRCA: Breast invasive carcinoma. HNSCC: Head and neck squamous cell carcinoma. ESCC: Esophageal carcinoma.).