A Database Tool Integrating Genomic and Pharmacologic Data from Adrenocortical Carcinoma Cell Lines, PDX, and Patient Samples

Abstract

Adrenocortical carcinoma (ACC) is a rare and highly heterogeneous disease with a notably poor prognosis due to significant challenges in diagnosis and treatment. Emphasizing on the importance of precision medicine, there is an increasing need for comprehensive genomic resources alongside well-developed experimental models to devise personalized therapeutic strategies. We present ACC_CellMinerCDB, a substantive genomic and drug sensitivity database (available at https://discover.nci.nih.gov/acc_cellminercdb) comprising ACC cell lines, patient-derived xenografts, surgical samples, and responses to more than 2,400 drugs examined by the NCI and National Center for Advancing Translational Sciences. This database exposes shared genomic pathways among ACC cell lines and surgical samples, thus authenticating the cell lines as research models. It also allows exploration of pertinent treatment markers such as MDR-1, SOAT1, MGMT, MMR, and SLFN11 and introduces the potential to repurpose agents like temozolomide for ACC therapy. ACC_CellMinerCDB provides the foundation for exploring larger preclinical ACC models.

Significance: ACC_CellMinerCDB, a comprehensive database of cell lines, patient-derived xenografts, surgical samples, and drug responses, reveals shared genomic pathways and treatment-relevant markers in ACC. This resource offers insights into potential therapeutic targets and the opportunity to repurpose existing drugs for ACC therapy.

Figure 1

Overview of the datasets and reproducibility of ACC CellMiner. A, URL and snapshot of the website for ACC_CellMinerCDB. B, Summary of the molecular and drug activity data for cell lines, patient-derived mouse xenografts, and surgical samples included in ACC CellMiner. For each type of molecular and drug data, numbers indicate how many genes or drugs are included. Gray boxes indicate items with no data. C, Table of samples overlapping between datasets (top) and cell lines included in each dataset (bottom). D, Distribution of gene expression correlation between “ACC NCI plus surgical” and “ACC Colorado plus PDX” data sets. E, CTNNB1, IGF2, and NR5A1 gene expressions in the two data sets are plotted, and Pearson’s correlation coefficients are shown at the top of the plots. *Only the TVBF-7 cell line has methylation data.

Figure 2

RNA-seq (xsq) expression of hormonal and drug efflux genes in the ACC cell lines and surgical samples. A, Univariate analysis scatterplot of the hormonal genes SULT2A1 (sulfotransferase family 1A member 1) vs. CYP11A1 (cytochrome P450 family 11 subfamily A member 1) transcript levels in ACC NCI cell lines and surgical samples. Transcript levels of the two genes in the 1,011 cell lines in the CCLE dataset were merged. B, Univariate analysis scatterplot of the drug transporters ABCB1 (MDR1) vs. ABCG2 (BCRP) transcript expression levels in ACC NCI cell lines and surgical samples. Transcriptional expression levels of ABCB1 and ABCG2 in the 1,011 cell lines of the CCLE dataset were merged with the ACC data.

Figure 3

Comparison of gene expression of CU-ACC cell lines and original PDXs. A, Overall correlation between gene expression in the cancer cell lines evaluated at the NCI and CU and their corresponding PDXs. Areas shaded in gray highlight the grouping of corresponding ACC cell lines and PDXs. B, Examples of coherent genes. Univariate analysis scatterplot of ABCB1 (MDR1) transcriptional expression level vs. ABCG2 (BCRP) and CYP21A2 transcriptional expression level vs. CYP11B1 transcriptional expression level. C, Example of differentially expressed genes related to cellular proliferation for the CU-ACC1 and CU-ACC2 cell lines and corresponding PDXs.

Figure 4

Examples of biomarker signatures in the ACC cell lines and surgical samples. A, The ADS varies across ACC cell lines and surgical samples and is correlated with the expression of LSS (lanosterol synthase), a key enzyme in cholesterol biosynthesis. B, The ADS is negatively correlated with the APM score.

Figure 5

Several ACC cell lines and surgical samples do not express TERT. A, Univariate scatterplot of TERT transcriptional expression levels vs. ATRX transcriptional expression levels in the ACC cell line and surgical sample data set. B, Alternative lengthening of telomeres (ALT)-associated PML body formation. Representative confocal micrograph images of ALT cell line (U2OS) and ACC cell lines. Cells were fixed and immunofluorescently labeled with Abs against TERF2 and PML; TERF2 was stained green, PML red, and nuclei DAPI blue. C, DAXX and ATRX mutations in Colorado, Zurich, and Wurzburg data sets. Mutation scores in each dataset were collected and plotted; MUC-1 and CU-ACC2 have ATRX mutations, and JIL-2266 has DXAA mutation.

Figure 6

Examples of gene copy number variations and mutations in the ACC cell lines. A, CU-ACC1 is defective in CDKN2A. Univariate scatterplot of CDKN2A transcriptional expression levels vs. CDKN2A gene copy number in the ACC NCI cell line data set. B, Gene mutations in the ACC cell lines; mutation scores were collected from the Colorado, Zurich, and Wurzburg datasets and plotted; CU-ACC2, MUC-1 and JIL-2266 are homozygous TP53 mutants, CU-ACC1 and H595R exhibit heterozygous β-catenin (CTNNB1) mutations. C, CU-ACC2 harbor homozygous NF1 gene mutation and heterozygous NF2 gene mutation in the ACC NCI cell line data set.

Figure 7

A, Correlation between the activity of mitotane tested at the NCI and the expression of SOAT1. B, Correlation between the activity of mitotane tested independently at the NCAT and the expression of SOAT1. C, Correlation between the expression of SOAT1 and the ADS signature.

Figure 8

The CU-ACC1 cell line is sensitive to temozolomide as it lacks MGMT and is MMR proficient, whereas CU-ACC2 cells are resistant to temozolomide because of MMR deficiency. A, CU-ACC1 and CU-ACC2 do not express MGMT transcripts. Univariate scatterplot of MGMT transcriptional expression levels vs. MGMT gene promoter methylation levels in the ACC cell line dataset. B, MGMT protein expression levels in CU-ACC1, CU-ACC2, NCI-H295R, and SW-13. Proteins were extracted from each cell line, and MGMT expression was assessed by Western blotting. C, The CU-ACC2 cell line is defective in MMR due to lack of expression of the MSH2 gene. Univariate scatter plot of MHS2 transcript levels vs. MSH2 gene copy number in the ACC cell lines data set. D, Dose–response curves of temozolomide in CU-ACC1, CU-ACC2, NCI-H295R, and SW-13. Cell viability was assessed after 72 hours under the indicated drug concentrations by CellTiter-Glo assay.