Printer Friendly

Molecular Portrait of Metastasis-Competent Circulating Tumor Cells in Colon Cancer Reveals the Crucial Role of Genes Regulating Energy Metabolism and DNA Repair.

Circulating tumor cells (CTCs) [9] are invasive tumor cells that disseminate from the primary tumor in the blood and might represent a fundamental prerequisite for metastasis formation (1). CTC analysis as a "real-time liquid biopsy" (1) in epithelial tumors (i.e., breast, prostate, lung, and colon cancer) for personalized medicine has received tremendous attention in recent years (2) and has important clinical implications (3). However, the low CTC concentration in blood samples, especially from patients with early stage cancer, is a crucial limiting factor for the identification of "metastasis-initiator cells." Moreover, information on CTC functional properties is quite limited because their low number in the peripheral blood makes difficult the generation and maintenance of in vitro cultures of CTCs from patients with tumors (4, 5). Recently, we derived the first cell line, named CTC-MCC-41, from CTCs isolated from the blood of a patient with colon cancer (6). CTC-MCC-41 cells have a stable intermediate epithelial-mesenchymal phenotype and stem cell--like properties, and these cells can produce tumors when xenografted in immunodeficient mice (6). CTC lines can be used to identify proteins and pathways involved in cancer dissemination and stemness and also to test new drugs to inhibit metastasis-competent CTCs. For these purposes, it is now important to determine the molecular bases underlying the functional differences between CTC lines, such CTC-MCC-41 cells, and cancer cell lines derived from the primary tumor.

In this study, we used DNA microarray technology to compare the transcriptome of metastasis-competent CTC-MCC-41 cells and of HT-29 cells, a cell line established from a primary colon cancer. This comparison allowed us to identify a specific molecular signature of CTCs that gives crucial information on their stem cell properties and their ability to initiate and support the formation of distant metastases. These findings may provide a starting point for investigating specific molecular mechanisms and key pathways involved in colon cancer progression. Moreover, such data may supply insights for the discovery of new biomarkers to identify the most aggressive CTC subpopulations and for the development of new drugs to inhibit metastasis-initiator CTCs.

Materials and Methods

CTC ISOLATION AND ESTABLISHMENT OF THE CTC-MCC-41 CELL LINE

We previously described the generation of the colon CTC line named CTC-MCC-41 (6) that now has been growing in culture for 3 years with a constant doubling time (see the Data Supplement that accompanies the online version of this article at http://www. clinchem.org/content/vol63/issue3).

CLINICOPATHOLOGIC CHARACTERISTICS OF THE PATIENT AND TUMOR GIVING RISE TO CTC-MCC-41

Standard histopathologic analysis of diagnostic biopsies and clinical data of the patient have been detailed in the online Supplemental Data.

HT-29 CELL LINE

The HT-29 cell line (ATCC[R]HTB-38[TM]) is a human colorectal adenocarcinoma cell line that was established in 1964 by J. Fogh from the primary tumor of a 44-yearold Caucasian female (7) (see online Supplemental Fig. S1).

RNA ISOLATION

The RNeasy Mini Kit (ref.74106; Qiagen) was used to extract total RNA from each cell sample, according to the manufacturers' recommended protocol (see online Supplemental Data). CTC-MCC-41 cells were prepared at passages P6, P8, and P10 and HT-29 cells at passages P130, P132, and P134.

COMPLEMENTARY RNA PREPARATION AND MICROARRAY HYBRIDATION

Total RNA (200 ng) was used to prepare cRNA using the Affymetrix 3' IVT express protocol (ref.901229). The complete cDNA was amplified by in vitro transcription (see online Supplemental Data). Amplified RNA (aRNA) was quantified with a NanoDrop ND1000 spectrophotometer. After fragmentation, 12 jig of labeled antisense aRNA was hybridized to HGU133 plus 2.0 GeneChip arrays (Affymetrix[R]). In total, 6 chips (3 chips for CTC-MCC-41 cell samples and 3 chips for HT-29 cell samples) were used for the microarray hybridization experiments. Microarray data were obtained and analyzed according to the minimal information about microarray experiment (MIAME) recommendations (8).

ACCESSION NUMBER

All our data are accessible at the gene expression Omnibus (GEO) repository (https://www.ncbi.nlm.nih.gov/geo) with the provisional accession series number GSE82198. The 14 publicly available Affymetrix data on other colon cancer cell lines are accessible through the accession numbers GSM224696, GSM224697, GSM224698, GSM274711, GSM274712, GSM1624316, GSM1624317, GSM274713, GSM274714, GSM800783, GSM274717, GSM274718, GSM274773, and GSM274774 (samples).

DATA PROCESSING AND BIOINFORMATICS ANALYSIS

After image processing with the Affymetrix GeneChip[R] Command Console[R] Software (AGCC), the CEL files were analyzed using the Affymetrix Expression Console[TM] software v1.3.1 and normalized with the MAS5.0 algorithm by scaling each array to a target value (TGT) of 100 using the global scaling method to obtain an intensity value signal for each probe set. First, microarray data were filtered based on the detection call (absent/present). Then, transcripts with significant differential expression profiles were identified using the 2-class Significance Analysis of Microarray (SAM) algorithm (http://www. stat.stanford.edu/~tibs/SAM/) with the Wilcoxon test and sample label permutations (n=300). Only transcripts with a significant false discovery rate (FDR) < 5% and fold change (FC) [greater than or equal to] 2 were retained. Hierarchical clustering analyses based on the expression levels of the differentially expressed genes were performed with the Cluster and TreeView software packages (9). Principal component analysis (PCA) was performed using GenomicScape (https://www.genomicscape. com). The Ingenuity Pathway Analysis (IPA) software (http://www.ingenuity.com) and Pathway Studio 9.0 (Ariadne Genomics) were used for the functional assessment of genes that were differentially expressed in CTC-MCC-41 and HT-29 cells (see online Supplemental Data). In addition, we used Affymetrix Transcriptome Analysis Console (TAC) software to construct a volcano plot and chromosome view.

REVERSE TRANSCRIPTION-QUANTITATIVE PCR ANALYSIS

Reverse transcription (RT) and quantitative PCR (qPCR) were performed as recommended by the manufacturer (Invitrogen) (see online Supplemental Data). The primer sequences are shown in the online Supplemental Table S1.

STATISTICAL ANALYSIS

Each experiment was performed as least 3 times and data are shown as the mean [+ or -] SE. The differences were evaluated using the Student t-test. Statistical analyses were performed with GraphPad Instat (Mann-Whitney Utest; GraphPad). A value of P [greater than or equal to] 0.05 was considered to be statistically significant.

Results

PHENOTYPIC AND MOLECULAR CHARACTERIZATION OF CTCMCC-41 CELLS (A METASTASIS-COMPETENT COLON CTC CELL LINE) AND HT-29 CELLS

Despite the different sources (circulating CTCs vs primary tumor) and culture methods (suspension vs adherent culture), the analysis of specific epithelial markers by immunocytochemistry revealed that both CTCMCC-41 and HT-29 cells expressed epithelial markers, such as EpCam and cytokeratin 20 (Fig. 1A). This suggests that these 2 colon cell lines have a similar phenotype. Then, the transcriptome profile of CTC-MCC-41 and HT29 cells was determined by using high-density oligonucleotide microarray analysis. The 3-dimensional distribution profile obtained by PCA of the variance for each gene showed that (a) samples from the same category (CTC-MCC-41 or HT29 cells) grouped together very tightly, corroborating the robustness of the microarray analysis, and that (b) CTC-MCC-41 cell samples were separated from HT-29 samples, demonstrating a significant difference in the molecular profile of these colon cancer cell lines (Fig. 1B). A volcano plot of the combined expression data of all transcripts grouped by fold change difference and P values (Fig. 1C) confirmed the results of the PCA analysis. The chromosomal distribution of transcripts expressed differently in CTCMCC-41 than in HT-29 cells are shown in a chromosome summary graph (see online Supplemental Fig. S2).

IDENTIFICATION OF SETS OF TRANSCRIPTS THAT ARE DIFFERENTIALLY EXPRESSED IN CTC-MCC-41 AND HT-29 CELLS

To gain insight into the molecular basis of the invasive properties of CTCs with stem cell properties, the transcriptome profiles of CTC-MCC-41 and HT-29 cells were then compared. A first selection based on the detection call retained 30167 transcripts. Then, the SAM software with a 2-fold change cutoff and FDR <5% identified 7553 differentially expressed transcripts (DETs) that significantly distinguished CTC-MCC-41 from H29 cell samples. Specifically, 3670 transcripts (probe sets) were upregulated in CTCMCC-41 cells and 3883 transcripts (probe sets) were upregulated in HT-29 cells (see online Supplemental Tables S2 and S3). Hierarchical clustering analysis based on the DETs clearly segregated CTC-MCC-41 from HT-29 cell samples (Fig. 2A). DETs were then functionally categorized using the IPA software. A large portion of DETs were involved in (a) cell morphology, cellular function, cellular assembly, and organization (score = 30) (Fig. 2B-a); (b) cell cycle, cellular movement, and cancer (score = 26) (Fig. 2B-b); and (c) metabolism, cell-to-cell signaling, and interaction (score = 26) (Fig. 2B-c). As expected, various blood cell- and cell movement-related adhesion molecules were found among the top regulator networks.

MOLECULAR DIVERGENCE BETWEEN CTC-MCC-41 AND HT-29 CELLS

The SAM analysis revealed that CTC-MCC-41 and HT-29 cells had specific molecular signatures. The "colon CTC signature" included 2324 genes (see online Supplemental Table S4) that were significantly upregulated in CTC-MCC-41 cells, among which 29% showed the highest FC (5 [less than or equal to] FC [less than or equal to] 1900). The "HT-29 signature" included 2294 genes (see online Supplemental Table S5) that were upregulated in HT-29 cells, among which 32% had the highest FC (5 [less than or equal to] FC [less than or equal to] 1102) (Fig. 3A). The "colon CTC signature" included the defensin [alpha] genes [defensin alpha 5 (DEFA5) and defensin alpha 6 (DEFA6)], [10] trefoil factors [trefoil factor 3 (TFF3) and trefoil factor 1 (TFF1)], an oncogene [KRAS protooncogene, GTPase (KRAS)], interleukins [interleukin-33 (IL-33), interleukin-37 (IL-37), and interleukin 12A (IL12A)], chemokines [C-X-Cmotifchemokineligand3 (CXCL3), C-X-C motif chemokine ligand 16 (CXCL16), and C-X-C motif chemokine ligand 24 (CCL24)], bone morphogenetic proteins [bone morphogenetic protein 5 (BMP5), bone morphogenetic protein 7 (BMP7), and bone morphogenetic protein 8B (BMP8B)], peroxisome proliferator-activated receptor [gamma] coactivators [peroxisome proliferator-activated receptor [gamma] coactivator 1A (PPARGC1A) and peroxisome proliferator-activated receptor [gamma] coactivator 1B (PPARGC1B)], and kallikrein-related peptidases [kallikrein 1 (KLK1), kallikrein 7 (KLK7), kallikrein 8 (KLK8), and kallikrein 15 (KLK15)]. Several genes encoding proteins involved in apoptosis were also significantly overexpressed in the "colon CTC signature," such as members of the tumor necrosis factor receptor superfamily [tumor necrosis factor receptor superfamily 1B (TNFRSF1B), tumor necrosis factor receptor superfamily 10B (TNFRSF10B), and tumor necrosis factor receptor superfamily 10A (TNFRSF10A)], apoptosis facilitators including BCL2 like 14 [BCL2L14 (BCL-G)], BCL2 like 15 [BCL2L15 (BFK)] and the BH3-only member BIM [BCL2 like 11 (BCL2L11)], apoptosis-inducers [BCL2 interacting killer (BIK)], death-associated protein kinase 2 (DAPK2), fas cell surface death receptor (FAS), and BCL2-associated X protein (BAX). Genes connected to p53 signaling, such as the genes encoding tumor protein 53 (TP53) inducible nuclear proteins [TP53 inducible nuclear protein 1 (TP53INP1) and ribonucleotide reductase regulatory TP53 inducible subunit M2B (RRM2B)], cyclins [cyclin D2 (CCND2) and cyclin D1 (CCND1)], MDM2 proto-oncogene (MDM2), pleomorphic adenoma gene-like 1 (PLAGL1), and DNA-damage regulated autophagy modulator 1 (DRAM1) were also upregulated in CTC-MCC-41 cells. Moreover, gremlin 1 (GREM1) and inhibin beta B subunit (INHBB) that encode ligands of the TGF-[beta] superfamily were among the most highly expressed transcripts in CTC-MCC-41 cells. Conversely, the "HT-29 signature" was enriched in genes associated with guidance and cytoskeleton signaling, such as platelet derived growth factor C (PDGFC), chromatin rearrangement genes [SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily A, member 1 (SMARCA1) and SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily A, member 2 (SMARCA2)], tissue inhibitor of metallopeptidases (TIMP) metallopeptidase inhibitor 2 (TIMP2), many ADAM metallopeptidases [ADAM metallopeptidase with thrombospondin type 1 motif 6 (ADAMTS6), ADAM metallopeptidase domain 28 (ADAM28), ADAM metallopeptidase domain 22 (ADAM22), ADAM metallopeptidase domain 19 (ADAM19), ADAM metallopeptidase domain 17 (ADAM17), and ADAM metallopeptidase domain 8 (ADAM8)], plexins [plexin D1 (PLXND1), plexin A2 (PLXNA2), and plexin A1 (PLXNA1)], integrins [integrin subunit alpha 1 (ITGA1), integrin subunit alpha 3 (ITGA3), integrin subunit alphaM (ITGAM), integrin subunit alpha V (ITGAV), integrin subunit beta 1 (ITGB1), integrin subunit beta 5 (ITGB5), and integrin subunit beta 8 (ITGB8)], and myosins [myosin light chain kinase (MYLK), myosin light chain 6 (MYL6), and myosin light chain 10 (MYH10)]. Interleukins [interleukin 8 (IL8), interleukin 6 signal transducer (IL6ST), interleukin 7 (IL7), interleukin 15 (IL15), and interleukin 1 receptor type 1 (IL1R1)] and chemokines [C-X-C motif chemokine ligand 2 (CXCL2), C-X-C motif chemokine ligand 6 (CXCL6), C-X-C motif chemokine ligand 10 (CXCL10), C-X-C motif chemokine ligand 11 (CXCL11), C-C motif chemokine ligand 5 (CCL5), and C-C motif chemokine ligand 28 (CCL28)] also were overexpressed in HT-29 cells. Finally, the Gene Set Enrichment Analysis showed that several tumorigenesis-related genes were enriched in the "colon CTC signature," such as carcinoembryonic antigen-related cell adhesion molecule 5 (CEACAM5) and carcinoembryonic antigen-related cell adhesion molecule 6 (CEACAM6), phospholipase A2 group 2A (PLA2G2A), and aldehyde dehydrogenase 3 family member A1 (ALDH3A1). Conversely, genes related to matrix metallopeptidases, such as prostaglandin-endoperoxide synthase 1 (PTGS1), elastin microfibril interfacer 2 (EMILIN2), procollagen-lysine 2-oxoglutarate 5-dioxygenase 2 (PLOD2), and collagen type 18 alpha 1 (COL18A1) were overrepresented in the "HT-29 signature" (Fig. 3, B and C).

MICROARRAY DATA VALIDATION BY RT-qPCR

Among all the DETs, 20 genes were selected for validation by RT-qPCR. These genes were chosen based either on their high FC or on their potential functions and involvement in a relevant biological pathway. Analysis of the qPCR data (Fig. 4) confirmed that DEFA6, B-cell CLL/lymphoma 11A (BCL11A), interleukin 33 (IL33), BMP7, ATP binding cassette subfamily B member 1 (ABCB1), CCND2, TNFRSF1B, fibronectin 1 (FN1), semaphorin 6A (SEMA6A), and galanin (GAL) were upregulated and that PDGFC, prostaglandin-endoperoxide synthase 2 (PTGS2), Wnt family member 11 (WNT11), transforming growth factor beta 2 (TGFB2), dickkopf WNT signaling pathway inhibitor 1 (DKK1), bone marrow stromal cell antigen 2 (BST2), GATA binding protein 2 (GATA2), gap junction protein beta 6 (GJB6), SMARCA1, and ADAM metallopeptidase with thrombospondin type 1 motif 6 (ADAMTS6) were downregulated in CTC-MCC-41 cells compared with HT-29 cells. This agreement between microarray and qPCRdata indicates that our microarray analysis was reliable.

MITOCHONDRIAL GENE EXPRESSION PROFILE AND METABOLIC PATHWAYS IN CTC-MCC-41 CELLS

Microarray data analysis also highlighted the upregulation of genes encoding proteins related to the "mitochondrial part" (GO: 0044429) and/or "mitochondrion" (GO: 0005739) GO categories in CTC-MCC-41 cells compared with HT-29 cells (Fig. 5A). The Pathway Studio analysis showed that the "colon CTC signature" was enriched in transcripts associated with localization in the "mitochondria" (Fig. 5B). The "colon CTC signature" also included genes known to promote mitochondrion biogenesis, particularly nuclear respiratory factor 1 (NRF1) and peroxisome proliferator-activated receptor [gamma] coactivators (PPARGC1A and PPARGC1B). Several metabolic genes were also upregulated in the "colon CTC signature," suggesting that different metabolic pathways are operating in metastasis-competent colon CTCs (Fig. 5C). Specifically, genes involved in fatty acid metabolism [fatty acid binding protein 1 (FABP1), fatty acid binding protein 5 (FABP5), ALDH3A1, aldehyde dehydrogenase 3 family member A2 (ALDH3A2), aldehyde dehydrogenase 1 family member B1 (ALDH1B1), and aldehyde dehydrogenase 2 (ALDH2)], estrogen biosynthesis [hydroxysteroid 17-beta dehydrogenase 1 (HSD17B1), cytochrome P450 family 2 subfamily B member 6 (CYP2B6), and aldo-keto reductase family 1 member C4 (AKR1C4)], PXR/RXR activation [glutathione S-transferase mu 1-4 (GSTM1, -2, -3, and -4)], energy production [ aldehyde oxidase 1 (AOX1), xanthine dehydrogenase (XDH), monoamine oxidase A (MAOA), monoamine oxidase B (MAOB), paraoxonase 2 (PON2), paraoxonase 3 (PON3), and voltage dependent anion channel 1 (VDAC1)], and various ATP-binding cassette (ABC) transporters (ABCB1, ABCA12, ABCA5, ABCA3, ABCC2, ABCC5, and ABCC6) were upregulated in CTC-MCC-41 cells. Finally, the gene encoding insulin-like growth factor binding protein 3 (IGFBP3) was strongly overexpressed in CTC-MCC-41 cells. This protein regulates apoptotic pathways by modulating the expression of BCL-2 family members in mitochondria (10-12).

EXPRESSION OF STEMNESS-RELATED GENES IN CTC-MCC-41 CELLS

Analysis of the expression of stemness-related genes in CTC-MCC-41 and HT-29 cells, based on a previously published data set from human embryonic stem cells (hESCs) in which a consensus hESC stemness gene list (n=1076 genes) was defined (13), showed that CTCMCC-41 cells shared 130 stemness-related genes with hESCs (see online Supplemental Table S6) and that HT-29 shared 109 genes. The key stemness genes glutaminase 2 (GLS2), cystathionine-beta-synthase (CBS), and cystathionine gamma-lyase (CTH) were enriched in CTC-MCC-41, whereas CD200 molecule (CD200), spalt like transcription factor 4 (SALL4), and inhibitor of DNA binding 1 (ID1) were over-represented in HT-29. Analysis of the functional relationship between the 130 stemness-related genes using IPA and Pathway Studio showed that 21 of them functionally interacted with each other, forming a tightly connected network, which we called "CTC core stemness circuitry" (Fig. 6A).

DNA REPAIR GENES IN CTC-MCC-41 CELLS

DNA repair pathways are crucial for the survival of cancer cells. The obtained microarray data were also used to evaluate the expression of a comprehensive list of 112 DNA repair genes (14) in CTC-MCC-41 and HT-29 cells (see online Supplemental Table S7). Among these genes, 9 [aprataxin (APTX), aprataxin and PNKP like factor (APLF), X-ray repair cross complementing 4 (XRCC4), BRCA1 interacting protein C-terminal helicase 1 (BRIP1), Fanconi anemia complementation group B (FANCB), Fanconi anemia complementation group M (FANCM), damage specific DNA binding protein 2 (DDB2), XPC complex subunit, DNA damage recognition and repair factor (XPC), and mutL homolog 3 (MLH3)] were over-expressed in CTC-MCC-41 cells and only 4 [DNA polymerase mu (POLM), RAD51 recombinase (RAD51), ERCC excision repair 1 (ERCC1), and tumor protein p53 binding protein 1 (TP53BP1)] in HT-29 cells. Analysis of the functional relationship between these DNA repair genes using the STRING software showed that they all displayed a documented functional interaction and that the Fanconi anemia (FA) pathway, which includes BRIP1, FANCB, and FANCM, was specifically upregulated in CTC-MCC-41 cells (Fig. 6B).

INTEGRATING AFFYMETRIX GENECHIP DATA SETS OBTAINED FROM OTHER COLON CANCER CELL LINES

To compare our data to those from other colon cancer cell lines, we used publicly available Affymetrix data on colon cell lines obtained from colon primary tumors (CaCo2, HCT116, and RKO) and colon cell lines obtained from metastatic distant sites: ascites and lymph node, respectively (C0L0205 and SW620). A hierarchical clustering analysis using Affymetrix TAC software mapped the samples into 2 main major clusters. As observed in the dendrogram (Fig. 6C), a first branch indicates a clear segregation between the CTC-MCC-41 and the other colon cancer cell lines (all the samples of the CTC-MCC-41 group self-cluster into 1 branch). The HT-29 samples self-cluster into another branch and situated in the vicinity of the primary and metastatic colon cancer cell lines. These results highlight that CTC-MCC-41 line displays a very specific transcription program. We further investigated differences and similarities in the gene expression patterns between CTC-MCC-41 lines and other colon cancer cell lines in comparison to HT-29 samples. A SAM, with a 2-fold change cutoff and FDR <5%, showed that 7380 transcripts were upregulated in the CaCo2 and HCT116 lines compared to HT-29 samples and 8924 transcripts in the C0L0205 and SW620 lines. These lists of transcripts were then intersected with the CTC-MCC-41 signature (see online Supplemental Table S4) to determine their overlap (Fig. 6D). Interestingly, among the 1624 transcripts (see online Supplemental Table S8) exclusively upregulated in CTC-MCC-41 samples, key genes related to energy metabolism such as (PPARGC1A, PPARGC1B, FABP1, ALDH3A, ALDH2, AOX1, and ABCB1), DNA repair (BRIP1, FANCM, XRCC4, DDB2, and XPC), and stemness genes including GLS2, ZNF93, and B3GNT7were observed.

Discussion

The analysis of the transcriptome profile of tumor cells is an important step for understanding cancer biology. In the present study, we compared the gene expression profiles of CTC-MCC-41 cells [derived from metastasis competent CTCs of a colon cancer patient (6) and of HT-29 cells (from a primary colorectal cancer)]. We clearly identified genes and signaling pathways that are involved in cancer progression and that were differentially expressed in CTC-MCC-41 and HT-29 cells. The genes upregulated in CTC-MCC-41 cells are involved in functions that are important for tumor cell survival, such as cell metabolism, cell growth and death, signal transduction, and DNA repair.

Tumor cells adapt the activity of metabolic pathways in response to modifications in the microenvironment (15) and need to coordinate energy generation to promote cell proliferation (16). Our data show that the expression levels of PPARGC1A (or PGCla) and PPARGC1B are much higher in CTC-MCC-41 than in HT-29 cells and that fatty acids metabolism is the most representative pathway in these colon CTCs. PPARGC1A is a major regulator of several key metabolic pathways and coordinates the regulation of genes promoting the conversion of glucose to fatty acids to support tumor growth (17). It also plays a pivotal role in inducing the expression of oxidative phosphorylation genes in various tissues, including colon cancer (18, 19). Increased fatty acid synthesis is important in cancer (20). PPARGC1A is a key modulator of the tumor suppressor p53 that promotes cell survival in response to metabolic stress (21) and can modulate PPARGC1 activity (22). Remarkably, many genes connected to p53 signaling were upregulated in CTC-MCC-41 cells, suggesting a potential response to stress caused by their presence in the bloodstream as well as a quick adaptation to the new microenvironment and a p53 role in the regulation of their metabolism.

The "CTC molecular signature" also included many genes that are annotated as "mitochondrial," demonstrating a strong link between CTCs and genes involved in energy production, cell proliferation, and apoptosis. Genes coding for BCL-2 family proteins are essential components of the intrinsic (mitochondrial) apoptotic pathway. The expression profile of apoptosis-related genes in CTC-MCC-41 cells suggests that the balance between anti- (BCL2A1) and pro-apoptotic factors (BIK, BAX, BCL2L11, BCL2L14, and BCL2L15) could be critical in CTCs. The antiapoptotic BCL2 gene is upregulated in many tumors, including colon and breast cancers (23-25). Moreover, the BIK protein heterodimerizes with BCL2 to antagonize its antiapoptotic function (26). Additionally, the expression level of TFF1, a protein involved in cell protection against several forms of apoptosis (27), was much higher in CTC-MCC-41 than in HT-29 cells. Thus, understanding apoptosis regulation in CTCs with stem cell properties should be critical for the success of therapies targeting these cells.

Moreover, some ABC transporter genes, such as ABCB1, ABCA12, ABCA5, ABCA3, ABCC2, ABCC5, and ABCC6, were strongly upregulated in CTC-MCC-41 compared with HT-29 cells. Several studies suggest that ABC transporters modulate cancer initiation and progression (28, 29). For instance, ABCA5 expression induction might correlate with the differentiation status of human colon cancers and contribute to tumor development (30) and ABCC2 transports a wide range of unconjugated organic anions (31). Our data are in agreement with previous studies reporting the upregulation of ABC transporters in colorectal cancer (28, 32).

In CTC-MCC-41 and HT-29 cells, multiple DNA repair pathways are active to repair genome damages and prevent genomic alterations. Our microarray analysis clearly shows that the BRIP1, FANCB, and FANCM genes, which are part of the FA pathway, are specifically upregulated in CTC-MCC-41 cells, suggesting that metastases-competent colon CTCs require the FA pathway for survival in vivo in the bloodstream and efficient growth in distant organs. BRIP1 alterations have been associated with the development of ovarian and breast cancers (33). However, the real function of FA pathway proteins in colon CTCs is not known yet. Similarly, many studies suggest that the FA pathway is especially important for pluripotent stem cells (PSCs) (34-37). Both PSCs and cancer cells can self-renew and proliferate, suggesting that the pathways controlling these bio logical processes might be shared between PSCs and cancer cells with metastasis-initiator properties. In this study, we used the stemness gene set previously identified by Assou et al. (13) in hESCs to analyze the "colon CTC signature." We found that stemness-related genes, including several genes that encode enzymes (GLS2, CBS/ CBSL, CTH, and GAD1), are preferentially upregulated in CTC-MCC-41 compared with HT-29 cells. Glutaminase 2 (GLS2) is a mitochondrial enzyme that converts glutamine into glutamate and is a p53 target gene that influences energy metabolism (38- 40). Cystathionine-[beta-synthase (CBS) and cystathionine-y-lyase (CTH) are the main enzymes that catalyze the production of hydrogen sulfide, which plays essential roles in many pathological processes (41). It was reported that colon cancer cells selectively over-express CBS to maintain the cell bioenergetics and to support tumor growth (42). Moreover, the WNT pathway regulates the expression of CTH, which plays important roles in colon cancer (43). Additional studies on the physiological function of these enzymes in CTC-MCC-41 cells will help better understand their role in colon cancer.

Another lesson from our transcriptomic analysis is that the CTC-MCC-41 lines display a very specific transcription program. Hierarchical clustering demonstrated that CTC-MCC-41 expression profiles were markedly different from those of HT-29 samples and other colon cancer cell lines (CaCo2, HCT116 and RKO, C0L0205, and SW620). In addition, the Venn diagram revealed that the differences observed between CTC-MCC-41 and HT-29 do not just reflect of differences between 2 different cell lines, but were exclusive to the CTC-MCC-41 metastasis competent cells. Additional studies of other yet-to-be-characterized metastasis competent cell lines will be necessary to determine whether these differences are unique to metastasis competent cell lines in general.

In conclusion, by focusing on the identified pathways and biomarkers, new drugs to target metastasis-competent CTC with stem cell properties could be developed to inhibit metastatic growth and relapses in patients with cancer. Many of the genes identified in this study may provide innovative biomarkers for CTC detection, isolation, and characterization. The new challenge is to correlate gene functions to specific events in colon CTCs. This may ultimately help in developing new personalized treatments for patients with colon cancer.

Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.

Authors' Disclosures or Potential Conflicts of Interest: Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest:

Employment or Leadership: None declared.

Consultant or Advisory Role: None declared.

Stock Ownership: None declared.

Honoraria: None declared.

Research Funding: C. Alix-Panabieres, DGOS and INCA grants and FEDER grant; T. Mazard, Roche (to the institution) and Amgen.

Expert Testimony: None declared.

Patents: None declared.

Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, and final approval of manuscript.

Acknowledgments: We are grateful to the Microarray Core Facility of the Institute for Regenerative Medicine and Biotherapy in Montpellier.

References

(1.) Pantel K, Alix-Panabieres C. Circulating tumour cells in cancer patients: challenges and perspectives. Trends Mol Med 2010;16:398-406.

(2.) Alix-Panabieres C, Pantel K. Challenges in circulating tumour cell research. Nat Rev Cancer 2014;14:623-31.

(3.) Alix-Panabieres C, Pantel K. Clinical applications of circulating tumor cells and circulating tumor DNA as liquid biopsy. Cancer Discov 2016;6:479-91.

(4.) Pantel K, Alix-Panabieres C. Functional studies on viable circulating tumor cells. Clin Chem 2016;62:328-34.

(5.) Pantel K, Alix-Panabieres C. Cell lines from circulating tumor cells. Oncoscience 2015;2:815-6.

(6.) Cayrefourcq L, Mazard T, Joosse S, Solassol J, Ramos J, Assenat E, et al. Establishment and characterization of a cell line from human circulating colon cancer cells. Cancer Res 2015;75:892-901.

(7.) Fogh J, Fogh JM, Orfeo T. One hundred and twenty-seven cultured human tumor cell lines producing tumors in nude mice. J Natl Cancer Inst 1977;59:221-6.

(8.) Brazma A, Hingamp P, Quackenbush J, Sherlock G, Spellman P, Stoeckert C, et al. Minimum information about a microarray experiment (MIAME)-toward standards for microarray data. Nat Genet 2001;29:365-71.

(9.) Eisen MB, Spellman PT, Brown PO, Botstein D. Cluster analysis and display of genome-wide expression patterns. Proc Natl Acad Sci USA1998;95:14863-8.

(10.) Lee KW, Ma L, Yan X, Liu B, Zhang XK, Cohen P. Rapid apoptosis induction by IGFBP-3 involves an insulin-like growth factor-independent nucleomitochondrial translocation of RXRalpha/Nur77. J Biol Chem 2005;280:16942-8.

(11.) Butt AJ, Firth SM, King MA, Baxter RC. Insulin-like growth factor-binding protein-3 modulates expression of Bax and Bcl-2 and potentiates p53-independent radiation-induced apoptosis in human breast cancer cells. J Biol Chem 2000;275:39174-81.

(12.) Toillon RA, Lagadec C, Page A, Chopin V, Sautiere PE, Ricort JM, et al. Proteomics demonstration that normal breast epithelial cells can induce apoptosis of breast cancer cells through insulin-like growth factor-binding protein-3 and maspin. Mol Cell Proteomics 2007;6:1239-47.

(13.) Assou S, Le Carrour T, Tondeur S, Strom S, Gabelle A, Marty S, et al. A meta-analysis of human embryonic stem cells transcriptome integrated into a web-based expression atlas. Stem Cells 2007;25:961-73.

(14.) Wood RD, Mitchell M, Lindahl T. Human DNA repair genes, 2005. Mutat Res 2005;577:275-83.

(15.) Shlomi T, Benyamini T, Gottlieb E, Sharan R, Ruppin E. Genome-scale metabolic modeling elucidates the role of proliferative adaptation in causing the Warburg effect. PLoS Comput Biol. 2011;7:e1002018.

(16.) Deberardinis RJ, Sayed N, Ditsworth D, Thompson CB. Brick by brick: metabolism and tumor cell growth. Curr Opin Genet Dev 2008;18:54-61.

(17.) Bhalla K, Hwang BJ, Dewi RE, Ou L, Twaddel W, Fang HB, etal. PGC1alpha promotes tumor growth by inducing gene expression programs supporting lipogenesis. Cancer Res 2011;71:6888-98.

(18.) Vellinga TT, Borovski T, de Boer VC, Fatrai S, van Schelven S,Trumpi K, et al. SIRT1/PGC1alpha-dependent increase in oxidative phosphorylation supports chemotherapy resistance of colon cancer. Clin Cancer Res 2015;21:2870-9.

(19.) Mootha VK, Lindgren CM, Eriksson KF, Subramanian A, Sihag S, Lehar J, et al. PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat Genet 2003; 34:267-73.

(20.) Menendez JA, Lupu R. Fatty acid synthase and the lipogenic phenotype in cancer pathogenesis. Nat Rev Cancer 2007;7:763-77.

(21.) Sen N, Satija YK, Das S. PGC-1alpha, a key modulator of p53, promotes cell survival upon metabolicstress. Mol Cell 2011;44:621-34.

(22.) Hallenborg P, Fjaere E, Liaset B, Petersen RK, Murano I, Sonne SB, et al. p53 regulates expression of uncoupling protein 1 through binding and repression of PPARgamma coactivator-1alpha. Am J Physiol Endocrinol Metab 2016;310:E116-28.

(23.) Bukholm IK, Nesland JM. Protein expression of p53, p21 (WAF1/CIP1), bcl-2, Bax, cyclin D1 and pRb in human colon carcinomas. Virchows Arch 2000;436:224-8.

(24.) Pappa G, Lichtenberg M, Iori R, Barillari J, Bartsch H, Gerhauser C. Comparison of growth inhibition profiles and mechanisms of apoptosis induction in human colon cancer cell lines by isothiocyanates and indoles from Brassicaceae. Mutat Res 2006;599:76-87.

(25.) Linjawi A, Kontogiannea M, Halwani F, Edwardes M, Meterissian S. Prognostic significance of p53, bcl-2, and Bax expression in early breast cancer. J Am Coll Surg 2004;198:83-90.

(26.) Han J, Sabbatini P, White E. Induction of apoptosis by human Nbk/Bik, a BH3-containing protein that interacts with E1B 19K. Mol Cell Biol 1996;16:5857-64.

(27.) Rodrigues S, Rodrigue CM, Attoub S, Flejou JF, Bruyneel E, Bracke M, et al. Induction of the adenomacarcinoma progression and Cdc25A-B phosphatases by the trefoil factor TFF1 in human colon epithelial cells. Oncogene 2006;25:6628-36.

(28.) Hlavata I, Mohelnikova-Duchonova B, Vaclavikova R, Liska V, Pitule P, Novak P, et al. The role of ABC transporters in progression and clinical outcome of colorectal cancer. Mutagenesis 2012;27:187-96.

(29.) Fletcher JI, Haber M, Henderson MJ, Norris MD. ABC transporters in cancer: more than just drug efflux pumps. Nat Rev Cancer 2010;10:147-56.

(30.) Ohtsuki S, Kamoi M, Watanabe Y, Suzuki H, Hori S, Terasaki T. Correlation of induction of ATP binding cassette transporter A5 (ABCA5) and ABCB1 mRNAs with differentiation state of human colon tumor. Biol Pharm Bull 2007;30:1144-6.

(31.) Haimeur A, Conseil G, Deeley RG, Cole SP. The MRP-related and BCRP/ABCG2 multidrug resistance proteins: biology, substrate specificity and regulation. Curr Drug Metab 2004;5:21-53.

(32.) Hinoshita E, Uchiumi T, Taguchi K, Kinukawa N, Tsuneyoshi M, Maehara Y, etal. Increased expression of an ATP-binding cassette superfamily transporter, multidrug resistance protein 2, in human colorectal carcinomas. Clin Cancer Res 2000;6:2401-7.

(33.) Ouhtit A, Gupta I, Shaikh Z. BRIP1, a potential candidate gene in development of non-BRCA1/2 breast cancer. Front Biosci (Elite Ed) 2016;8:289-98.

(34.) Chlon TM, Hoskins EE, Mayhew CN, Wikenheiser-Brokamp KA, Davies SM, Mehta P, et al. High-risk human papillomavirus E6 protein promotes reprogramming of Fanconi anemia patient cells through repression of p53 but does not allow for sustained growth of induced pluripotentstem cells. J Virol 2014; 88:11315-26.

(35.) Muller LU, Milsom MD, Harris CE, Vyas R, Brumme KM, ParmarK, etal. Overcoming reprogramming resistance of Fanconi anemia cells. Blood 2012;119:5449-57.

(36.) Liu GH, Suzuki K, Li M, Qu J, Montserrat N, Tarantino C, et al. Modelling Fanconi anemia pathogenesis and therapeutics using integration-free patient-derived iPSCs. Nat Commun 2014;5:4330.

(37.) Chlon TM, Ruiz-Torres S, Maag L, Mayhew CN, Wikenheiser-Brokamp KA, Davies SM, et al. Overcoming Pluripotent Stem Cell Dependence on the Repair of Endogenous DNA Damage. Stem Cell Reports 2016;6:44-54.

(38.) Hu W,Zhang C, Wu R,Sun Y, LevineA, Feng Z. Glutaminase 2, a novel p53 targetgene regulating energy metabolism and antioxidant function. Proc Natl Acad Sci U SA2010;107:7455-60.

(39.) Suzuki S, Tanaka T, Poyurovsky MV, Nagano H, Mayama T, Ohkubo S, et al. Phosphate-activated glutaminase (GLS2), a p53-inducible regulator of glutamine metabolism and reactive oxygen species. Proc Natl Acad Sci U SA2010;107:7461-6.

(40.) Giacobbe A, Bongiorno-Borbone L, Bernassola F, Terrinoni A, Markert EK, Levine AJ, etal. p63 regulates glutaminase 2 expression. Cell Cycle 2013;12:1395-405.

(41.) Szabo C. Hydrogen sulphide and its therapeutic potential. Nat Rev Drug Discov 2007;6:917-35.

(42.) Szabo C, Coletta C, Chao C, Modis K, Szczesny B, Papapetropoulos A, et al. Tumor-derived hydrogen sulfide, produced by cystathionine-beta-synthase, stimulates bioenergetics, cell proliferation, and angiogenesis in colon cancer. Proc Natl Acad Sci U S A 2013;110: 12474-9.

(43.) Fan K, Li N, Qi J, Yin P, Zhao C, Wang L, et al. Wnt/ beta-catenin signaling induces the transcription of cystathionine-gamma-lyase, a stimulator of tumor in colon cancer. Cell Signal 2014;26:2801-8.

Catherine Alix-Panabieres, [1,2] * Laure Cayrefourcq, [1,2] Thibault Mazard, [3,4] Thierry Maudelonde, [5,2] Eric Assenat, [6] and Said Assou [7,8] *

[1] Laboratory of Rare Human Circulating Cells, Department of Cellular and Tissue Biopathology of Tumors, University Medical Centre, Montpellier, France; [2] EA2415--Help for Personalized Decision: Methodological Aspects, University Institute of Clinical Research (IURC), University of Montpellier, Montpellier, France; [3] Department of Medical Oncology, Institut du Cancer a Montpellier (ICM), France; [4] Institut du Cancer a Montpellier (ICM), Montpellier, France; [5] Laboratory of Hormonal and Cell Biology, University Medical Centre, Montpellier, France; [6] Department of Medical Oncology, University Medical Centre, Montpellier, France; [7] University of Montpellier, UFRde Medecine, Montpellier, France; [8] INSERM U1183; Institute for Regenerative Medicine and Biotherapy, CHU Montpellier, Hopital Saint-Eloi, Montpellier, France.

* Address correspondence to C.A.-P. at Institut Universitairede Recherche Clinique (IURC), 641, Ave. du Doyen Gaston Giraud, 34093 Montpellier Cedex 5, France. Fax +33-4-1175-99-33; e-mail c-panabieres@chu-montpellier.fr. S.A. at Institute for Research in Regenerative Medicine and Biotherapy, Hopital Saint-Eloi, 80 Ave. Augustin Fliche, 34295 Montpellier Cedex 5, France. Fax +33-4-67-33-01-13; e-mail said.assou@inserm.fr.

Received July 11, 2016; accepted October 24, 2016.

Previously published online at DOI: 10.1373/clinchem.2016.263582

[9] Nonstandard abbreviations: CTCs, circulating tumor cells; MIAME, minimal information about microarray experiment; GEO, gene expression Omnibus; AGCC, Affymetrix GeneChip Command Console software; SAM, significance analysis of microarray; FDR, false discovery rate; FC, fold change; PCA, principal component analysis; IPA, ingenuity pathway analysis; TAC, transcriptome analysis console; RT, reverse transcription; qPCR, quantitative PCR; DETs, differentially expressed transcripts; hESCs, human embryonic stem cells; FA, fanconi anemia; PSCs, pluripotentstem cells; GLS2, glutaminase 2; CBS, cystathionine-,8-synthase; CTH, cystathionine-y-lyase.

[10] Human genes: DEFA5, defensin alpha 5; DEFA6, defensin alpha 6; TFF3, trefoil factor 3; TFF1, trefoil factor 1; KRAS, KRAS proto-oncogene, GTPase; IL-33, interleukin-33; IL-37, interleukin-37; IL12A, interleukin 12A; CXCL3, C-X-C motif chemokine ligand 3; CXCL16, C-X-C motif chemokine ligand 16; CCL24, C-X-C motif chemokine ligand 24; BMP5, bone morphogenetic protein 5; BMP7, bone morphogenetic protein 7; BMP8B, bone morphogenetic protein 8B; PPARGC1A, peroxisome proliferator-activated receptor y coactivator 1A; PPARGC1B, peroxisome proliferator-activated receptor y reactivator 1B; KLK1, kallikrein 1; KLK7, kallikrein 7; KLK8, kallikrein 8; KLK15, kallikrein 15; TNFRSF1B, tumor necrosis factor receptor superfamily 1B; TNFRSF10B, tumor necrosis factor receptorsuperfamily 10B; TNFRSF10A, tumor necrosis factor receptor superfamily 10A; BCL2L14 (BCL-G), BCL2 like 14; BCL2L15 (BFK), BCL2 like 15; BCL2L11, BCL2 like 11; BIK, BCL2 interacting killer; DAPK2, death-associated protein kinase 2; FAS, fas cell surface death receptor; BAX, BCL2-associated X protein; TP53, tumor protein 53; TP53INP1, TP53 inducible nuclear protein 1; RRM2B, ribonucleotide reductase regulatory TP53 induciblesubunit M2B; CCND2, cyclin D2; CCND1, cyclin D1; MDM2, MDM2 proto-oncogene; PLAGL1, pleomorphic adenoma gene-like 1; DRAM1, DNA-damage regulated autophagy modulator 1; GREM1, gremlin 1; INHBB, inhibin beta B subunit; PDGFC, platelet derived growth factor C; SMARCA1, SWI/SNF related, matrix-associated, actin dependent regulator of chromatin, subfamily A, member 1; SMARCA2, SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily A, member 2; TIMP, tissue inhibitor of metallopeptidases; TIMP2, metallopeptidase inhibitor 2; ADAMTS6, ADAM metallopeptidase with thrombospondin type 1 motif 6; ADAM28, ADAM metallopeptidase domain 28; ADAM22, ADAM metallopeptidase domain 22; ADAM19, ADAM metallopeptidase domain 19; ADAM17, ADAM metallopeptidase domain 17; ADAM8, ADAM metallopeptidase domain 8; PLXND1, plexin D1; PLXNA2, plexin A2; PLXNA1, plexin A1; ITGA1, integrin subunit alpha 1; ITGA3, integrin subunit alpha 3; ITGAM, integrin subunit alpha M; ITGAV, integrin subunit alpha V; ITGB1, integrinsubunit beta 1; ITGB5, integrin subunit beta 5; ITGB8, integrin subunit beta 8; MYLK, myosin light chain kinase; MYL6, myosin light chain 6; MYH10, myosin light chain 10; IL8, interleukin 8; IL6ST, interleukin 6 signal transducer; IL7, interleukin 7; IL15, interleukin 15; IL1R1, interleukin 1 receptortype 1; CXCL2, C-X-Cmotifchemokine ligand 2; CXCL6, C-X-C motif chemokine ligand 6; CXCL10, C-X-C motif chemokine ligand 10; CXCL11, C-X-C motif chemokine ligand 11; CCL5, C-C motif chemokine ligand 5; CCL28, C-C motif chemokine ligand 28; CEACAM5, carcinoembryonic antigen-related cell adhesion molecule 5; CEACAM6, carcinoembryonic antigen-related cell adhesion molecule 6; PLA2G2A, phospholipase A2 group 2A; ALDH3A1, aldehyde dehydrogenase 3 family member A1; PTGS1, prostaglandin-endoperoxide synthase 1; EMILIN2, elastin microfibril interfacer 2; PLOD2, procollagen-lysine 2-oxoglutarate 5-dioxygenase 2; COL18A1, collagen type 18 alpha 1; BCL11A, B-cell CLL/lymphoma 11A; IL33, interleukin 33; ABCB1, ATP binding cassette subfamily B member 1; FN1, fibronectin 1; SEMA6A, semaphorin 6A; GAL, galanin; PTGS2, prostagland in endoperoxide synthase 2; WNT11, Wnt family member 11; TGFB2, transforming growth factor beta 2; DKK1, dickkopf WNT signaling pathway inhibitor 1; BST2, bone marrow stromal cell antigen 2; GATA2, GATA binding protein 2; GJB6, gap junction protein beta 6; ADAMTS6, ADAM metallopeptidase with thrombospondin type 1 motif 6; NRF1, nuclear respiratory factor 1; FABP1, fatty acid binding protein 1; FABP5, fatty acid binding protein 5; ALDH3A2, aldehyde dehydrogenase 3 family member A2; ALDH1B1, aldehyde dehydrogenase 1 family member B1; ALDH2, aldehyde dehydrogenase 2; HSD17B1, hydroxysteroid 17-beta dehydrogenase 1; CYP2B6, cytochrome P450 family 2 subfamily B member 6; AKR1C4, aldo-keto reductase family 1 member C4; GSTM1,-2,-3, and -4, glutathione S-transferase mu 1-4; AOX1, aldehyde oxidase 1; XDH, xanthine dehydrogenase; MAOA, monoamine oxidase A; MAOB, monoamine oxidase B; PON2, paraoxonase2; PON3, paraoxonase3; VDAC1, voltage dependentanion channel 1; ABC, ATP-binding cassette; IGFBP3, insulin-like growth factor binding protein 3; GLS2, glutaminase 2; CBS, cystathionine-beta-synthase; CIH, cystathionine gamma-lyase; CD200, CD200 molecule; SALL4, spalt like transcription factor 4; ID1, inhibitor of DNA binding 1; APTX, aprataxin; APLF, aprataxin and PNKP like factor; XRCC4, X-ray repair cross complementing 4; BRIP1, BRCA1 interacting protein C-terminal helicase 1; FANCB, Fanconi anemia complementation group B; FANCM, Fanconi anemia complementation group M; XPC, XPC complexsubunit, DNA damage recognition and repair factor; DDB2, damage specific DNA binding protein 2; MLH3, mutL homolog 3; POLM, DNA polymerase mu; RAD5T, (RAD51 recombinase; ERCC1, ERCC excision repair 1; TP53BP1, tumor protein p53 binding protein 1.

Caption: Fig. 1. Characterization of CTC-MCC-41 and HT-29 cells. (A), Representative images of single CTC-MCC-41 cells and spheres and HT-29 cells obtained using a bright-field microscope (magnification, x40) (upper panels). Immunocytochemical staining of CTC-MCC-41 and HT-29 cells with DAPI, anti-EpCAM-APC, anti-CK20- FITC, and antiCD45-PE antibodies (lower panels). (B), PCA 3-dimensional (Dim.) scatter plots representing the gene expression patterns of the different samples (CTC-MCC-41 and HT-29 cells). Each dot represents a sample and the symbol its origin: C, CTC-MCC-41 cells; H, HT-29 cells. Samples can be divided in 2 distinct groups based on their gene expression profiles. (C), Volcano plotsshowing the distribution of gene expression fold changes and P values. A total of 54675 transcripts were tested using the TAC software. Transcripts upregulated in CTC- MCC-41 are indicated in red, and transcripts upregulated in HT-29 are indicated in green.

Caption: Fig. 2. CTC-MCC-41 and HT-29 gene signatures. (A), Heat map. The molecular signatures of CTC-MCC-41 and HT-29 cells were visualized by hierarchical clustering based on the DETs. Genes are arranged in rows and the 6 cell samples are arranged in columns. All 3 CTC-MCC-41 cell samples clustered in 1 branch and all HT-29 cell samples in another branch. In each cell group, the tree represents the relationship among samples and the branch length reflects the degree of similarity between samples according to their gene expression profile. Red, upregulated genes; green, downregulated genes. (B), Functional network analyses based on the DETs. The most significant networks included genes related to: (a) cell morphology, cell function, cell assembly, and organization; (b) cell cycle, cell movement, and cancer; and (c) metabolism, cell-to-cell signaling, and interaction. Upregulated genes are in pink and downregulated genes are in green. Continuous lines between nodes indicate direct molecular interactions between connected genes and dotted lines indicate indirect functional interactions between genes.

Caption: Fig. 3. Differentially expressed genes in CTC-MCC-41 cells and HT-29 cells. (A), Number of upregulated genes in CTC-MCC-41 cells and in HT-29 cells. SAM analysis identified a signature of 2324 genes that are upregulated in CTC-MCC-41 cells and 2294 genes that are upregulated in HT-29 cells. Of these, 46 and 38 genes showed a fold change [greater than or equal to] 100 in CTC-MCC-41 cells and HT-29 cells, respectively. (B), Gene Set Enrichment Analysis (GSEA) was performed using the microarray data to compare the gene expression profiles of CTC-MCC-41 and HT-29 cells. Upper panels present GSEA results showing a significant enrichment of the tumorigenesis category among genes upregulated in CTC-MCC-41 cells. Lower panels present GSEA results showing that the MMP14 signaling pathway is downregulated in CTC-MCC-41 cells compared with HT-29 cells. The heatmaps on the right show the gene expression level (red, high; blue, low) of each gene in the indicated cell sample. (C), Box-and-whisker plots comparing the expression level of DETs in CTC-MCC-41 and HT-29 cell samples. The signal intensity for each gene is shown on the y axis as arbitrary units determined by the AGCC software (Affymetrix).

Caption: Fig. 4. Validation by RT-qPCR of the microarray data. A set (n=20) of upregulated and downregulated genes in CTC-MCC-41 cells were analyzed by RT-qPCR to validate the microarray data. All the RT-qPCR results were normalized to the expression level of [beta]2-microglobulin in each sample. The relative abundance for each gene is shown on the y axis in arbitrary units. Results are presented as the mean [+ or -] SE. *P value <0.05.

Caption: Fig. 5. Expression of mitochondrial-related genes in CTC-MCC-41 cells and HT-29 cells. (A), Histograms show the percentage of genes annotated as "mitochondrial part" and "mitochondrion" (GO categories) among the genes overexpressed in CTC-MCC-41 (dark grey) and in HT-29 (light grey) cells. (B), Pathway studio analysis of differentially expressed genes in CTC-MCC-41 (CTC) and HT-29 cells. Only genes included in the "colon CTC signature" are connected to mitochondria. Each node represents either a gene entity or a control mechanism of the interaction. (C), Top canonical pathways that were significantly upregulated in CTC-MCC-41 cells, identified by Ingenuity[R] Pathway Analysis (IPA). The "fatty acid metabolism" and "xenobiotic signaling" pathways were significantly associated with genes overexpressed in CTC-MCC-41 cells. The ratio represents the number of analyzed genes in a given pathway divided by the total number of genes that form that pathway. The threshold represents the P value.

Fig. 6. Core stemness circuitry and DNA repair genes. (A), Functional networks for stemness-related genes, identified using Pathway Studio, in CTC-MCC-41 cells. Interactions between nodes are denoted by colored arrows or lines. The 21 stemness related-genes included in the "colon CTC signature" are listed in the table. (B), DNA repair genes in CTC-MCC-41 cells and HT-29 cells. This figure was produced using the STRING software to generate automatically the interaction graph of DNA repair genes. Colored circles represent genes, and lines indicate interactions among genes. (C), Hierarchical clustering. Each horizontal line represents a transcript and each column represents a single sample. A tree represents relationship among samples whose branch lengths reflect the degree of similarity between the samples according to transcript expression profile. Transcripts up- and down-regulated were colored in red and blue, respectively. (D), Intersection with transcriptomes of colon cancer cell lines. Venn diagram representing the number of genes in each comparison and the overlaps between the 3 main comparison groups.
Symbol     Entrez Gene Name                 Type(s)

BAX        BCL2-associated X protein        transporter

CALB1      calbindin 1                      other

CBS/CBSL   cystathionine-beta-synthase      enzyme

CTH        cystathionine gamma-lyase        enzyme

EGLN3      egl-9 family hypoxia-inducible   enzyme
           factor 3

EIF2S1     eukaryotic translation           translation regulator
           initiation factor 2

EIF4BP1    eukaryotic translation           other
           initiation factor 4B
           pseudogene 1

GAD1       glutamate decarboxylase 1        enzyme

GAL        galanin/GMAP prepropeptide       other

GLS2       glutaminase 2                    enzyme

HSPA9      heat shock 70kDa protein 9       other
           (mortalin)

HSPD1      heat shock 60kDa protein 1       enzyme
           (chaperonin)

ITGA6      integrin, alpha 6                transmembrane receptor

KCNN2      potassium channel                ion channel

LCK        LCK proto-oncogene, Src family   kinase
           tyrosine kinase

MMP1       matrix metallopeptidase 1        peptidase

MYB        v-myb avian myeloblastosis       transcription regulator
           viral oncogene homolog

NPTX2      neuronal pentraxin II            other

PPP2R1B    protein phosphatase 2,           phosphatase
           regulatory subunit A, beta

SEMA3A     semaphorin 3A                    other

VSNL1      visinin-like 1                   other
COPYRIGHT 2017 American Association for Clinical Chemistry, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Cancer Diagnostics
Author:Alix-Panabieres, Catherine; Cayrefourcq, Laure; Mazard, Thibault; Maudelonde, Thierry; Assenat, Eric
Publication:Clinical Chemistry
Article Type:Report
Date:Mar 1, 2017
Words:7582
Previous Article:Patient-Specific Circulating Tumor DNA Detection during Neoadjuvant Chemotherapy in Triple-Negative Breast Cancer.
Next Article:Two-Stage Isothermal Enzymatic Amplification for Concurrent Multiplex Molecular Detection.
Topics:

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters