Detection of MET Gene Copy Number in Cancer Samples Using the Droplet Digital PCR Method

Purpose

The analysis of MET gene copy number (CN) has been considered to be a potential biomarker to predict the response to MET-targeted therapies in various cancers. However, the current standard methods to determine MET CN are SNP 6.0 in the genomic DNA of cancer cell lines and fluorescence in situ hybridization (FISH) in tumor models, respectively, which are costly and require advanced technical skills and result in relatively subjective judgments. Therefore, we employed a novel method, droplet digital PCR (ddPCR), to determine the MET gene copy number with high accuracy and precision.

Methods

The genomic DNA of cancer cell lines or tumor models were tested and compared with the MET gene CN and MET/CEN-7 ratio determined by SNP 6.0 and FISH, respectively.

Results

In cell lines, the linear association of the MET CN detected by ddPCR and SNP 6.0 is strong (Pearson correlation = 0.867). In tumor models, the MET CN detected by ddPCR was significantly different between the MET gene amplification and non-amplification groups according to FISH (mean: 15.4 vs 2.1; P = 0.044). Given that MET gene amplification is defined as MET CN >5.5 by ddPCR, the concordance rate between ddPCR and FISH was 98.0%, and Cohen''s kappa coefficient was 0.760 (95% CI, 0.498–1.000; P <0.001).

Conclusions

The results demonstrated that the ddPCR method has the potential to quantify the MET gene copy number with high precision and accuracy as compared with the results from SNP 6.0 and FISH in cancer cell lines and tumor samples, respectively.     

Relationships between Chromosome 7 Gain,MET Gene Copy Number Increase and MET Protein Overexpression in Chinese Papillary Renal Cell Carcinoma Patients

To investigate the relationships between Chromosome 7 gain, mesenchymal-epithelial transition factor (MET) gene copy number increase and MET protein overexpression in Chinese patients with papillary renal cell carcinoma (PRCC), immunohistochemistry (IHC), immunofluorescence (IF) and fluorescence in situ hybridization (FISH) were performed on 98 formalin-fixed, paraffin-embedded (FFPE) PRCC samples. Correlations between MET gene copy number increase, Chromosome 7 gain and MET protein overexpression were analyzed statistically. A highly significant correlation was observed between the percentage of tumor cells with MET gene copy number ≥3 and CEP7 copy number ≥3 (R² = 0.90, p<0.001) across two subtypes of PRCC. In addition, the percentage of tumor cells with MET gene copy number ≥3 was found to increase along with increases in MET IHC score. This correlation was further confirmed in those PRCC tumor cells with average MET gene copy number >5 using combined IF and FISH methodology. Overall, this study provides evidence that Chromosome 7 gain drives MET gene copy number increase in PRCC tumors, and appears to subsequently lead to an increase in MET protein overexpression in these tumor cells. This supports MET activation as a potential therapeutic target in sporadic PRCC.     

Localization of amplified CAD genes on rearranged chromosomes of Chinese hamster cells

Chinese hamster cell lines carrying an amplified CAD region were selected from V79,B7 cells by their resistance to N-phosphonacetyl-L-aspartate (PALA). In one of the selected cell lines, SP PALA _inf1 ^supR L, an acrocentric chromosome with abnormally elongated q arms was identified as a marker for the PALA-resistant phenotype. The marker chromosome carried a homogeneously staining region close to a telomeric nucleolar organizer region. In the same region, localization of amplified CAD sequences was demonstrated by in situ hybridization. The marker chromosome was found to undergo extensive rearrangements. In particular, dicentric chromosomes, occurring with an unusually high incidence, were found to originate from end-fusion of two homologous marker chromosomes.Abbreviations ATCase Aspartate Transcarbamylase - CAD Carbamyl-phosphate synthetase-Aspartate transcarbamylase-Dihydroorotase - MTX Methotrexate - NOR Nucleolar Organizer Region - PALA N-phosphonacetyl-L-Aspartate     

MET Gene Amplification and MET Receptor Activation Are Not Sufficient to Predict Efficacy of Combined MET and EGFR Inhibitors in EGFR TKI-Resistant NSCLC Cells

Epidermal growth factor receptor (EGFR), member of the human epidermal growth factor receptor (HER) family, plays a critical role in regulating multiple cellular processes including proliferation, differentiation, cell migration and cell survival. Deregulation of the EGFR signaling has been found to be associated with the development of a variety of human malignancies including lung, breast, and ovarian cancers, making inhibition of EGFR the most promising molecular targeted therapy developed in the past decade against cancer. Human non small cell lung cancers (NSCLC) with activating mutations in the EGFR gene frequently experience significant tumor regression when treated with EGFR tyrosine kinase inhibitors (TKIs), although acquired resistance invariably develops. Resistance to TKI treatments has been associated to secondary mutations in the EGFR gene or to activation of additional bypass signaling pathways including the ones mediated by receptor tyrosine kinases, Fas receptor and NF-kB. In more than 30–40% of cases, however, the mechanisms underpinning drug-resistance are still unknown. The establishment of cellular and mouse models can facilitate the unveiling of mechanisms leading to drug-resistance and the development or validation of novel therapeutic strategies aimed at overcoming resistance and enhancing outcomes in NSCLC patients. Here we describe the establishment and characterization of EGFR TKI-resistant NSCLC cell lines and a pilot study on the effects of a combined MET and EGFR inhibitors treatment. The characterization of the erlotinib-resistant cell lines confirmed the association of EGFR TKI resistance with loss of EGFR gene amplification and/or AXL overexpression and/or MET gene amplification and MET receptor activation. These cellular models can be instrumental to further investigate the signaling pathways associated to EGFR TKI-resistance. Finally the drugs combination pilot study shows that MET gene amplification and MET receptor activation are not sufficient to predict a positive response of NSCLC cells to a cocktail of MET and EGFR inhibitors and highlights the importance of identifying more reliable biomarkers to predict the efficacy of treatments in NSCLC patients resistant to EGFR TKI.     

Duplication of theDR3Gene on Human Chromosome 1p36 and Its Deletion in Human Neuroblastoma

The humanDR3gene, whose product is also known as Wsl-1/APO-3/TRAMP/LARD, encodes a tumor necrosis factor-related receptor that is expressed primarily on the surface of thymocytes and lymphocytes. DR3 is capable of inducing both NF-κB activation and apoptosis when overexpressed in mammalian cells, although its ligand has not yet been identified. We report here that theDR3gene locus is tandemly duplicated on human chromosome band 1p36.2–p36.3 and that these genes are hemizygously deleted and/or translocated to another chromosome in neuroblastoma (NB) cell lines with amplifiedMYCN.Duplication of at least a portion of theDR3gene, including the extracellular and transmembrane regions but not the cytoplasmic domain, was demonstrated by both fluorescencein situhybridization and genomic Southern blotting. In most NB cell lines, both theDR3and theDR3Lsequences are simultaneously deleted and/or translocated to another chromosome. Finally, DR3/Wsl-1 protein expression is quite variable among these NB cell lines, with very low or undetectable levels in 7 of 17 NB cell lines.     

Cytogenetic rearrangements involving the loss of the Sonic Hedgehog gene at 7q36 cause holoprosencephaly

Holoprosencephaly (HPE) is a genetically heterogeneous disorder that affects the midline development of the forebrain and midface in humans. As a step toward identifying one of the HPE genes, we have set out to refine the HPE3 critical region on human chromosome 7q36 by analyzing 34 cell lines from families with cytogenetic abnormalities involving 7q, 24 of which are associated with HPE. Genomic clones surrounding the DNA marker D7S104, which has previously been shown to be in the HPE3 critical region, have been examined by fluorescent in situ hybridization and microsatellite analysis of our panel of patient cell lines. We report the analysis of a cluster of four translocation breakpoints within a 300-kb region of 7q36 that serves to define the minimal critical region for HPE3 and that has directed the search for candidate genes. The human Sonic Hedgehog (hSHH) gene maps to this region and has been shown to be HPE3 on the basis of mutations within the coding region of the gene. We present evidence that cytogenetic deletions and/or rearrangements of this region of chromosome 7q containing Sonic Hedgehog, and translocations that may suppress Sonic Hedgehog gene expression through a position effect are common mechanisms leading to HPE. Received: 23 December 1996 / Accepted: 17 March 1997     

Crossovers in two German cystic fibrosis families determine probe order for MET, 7C22 and XV-2c/CS.7

Summary We have followed the segregation of the probes pJ3.11, 7C22, pB79a, and MET through cystic fibrosis families in the German Democratic Republic with two affected sibs. Two families with a crossover between MET and the CF phenotype were detected. In one of these families recombination was also observed between the DNA probe 7C22 and CF, and between the markers XV-2c and CF, which suggests that XV-2c, MET and 7C22 are all on the same side of CF. The other MET recombinant family is informative with XV-2c and does not recombine, which excludes the genetic order XC-2c-MET-CF if multiple recombinant events are disregarded. These two families together demonstrate that recombinations may occur in a very small genetic interval, which has important implications for prenatal diagnosis based on data from linked markers.     

Genetic characterization of parental cell lines and biochemical analysis of somatic cell hybrids between mouse (RAG) cells and fibroblasts of Ateles paniscus chamek (primates,platyrrhini)

Mouse (RAG) cells, (deficient in hypoxanthine-phosphoribosyl-transferase), and Ateles paniscus chamek primary fibroblasts were used in fusion experiments to generate somatic cell hybrids. Both parental cell lines were genetically characterized by karyological and biochemical analyses with 27 isozyme systems. These procedures were useful for monitoring primate chromosome segregation in somatic cell hybrids, for detecting chromosome rearrangements of primate chromosomes, and for identifying individual primate chromosomes. These characterizations are necessary to distinguish between different hybrid cell lines and to generate a panel for gene mapping studies. This is achieved by selecting cell lines that segregate different sets of relatively few primate isozymes and chromosomes. Conversely, we eliminated hybrid cell lines either showing: (1) rearrangements between primate and mouse chromosomes, (2) extensive rearrangements of primate chromosomes, or (3) a large number of primate biochemical markers. © 1993 Wiley-Liss, Inc.     

Isolation of a polymorphic genomic clone from chromosome 7

Summary A peptide prepared from purified factor 13B (F13B) was sequenced, and a single, long oligonucleotide corresponding to its cognate DNA sequence was constructed and used to screen a chromosome 7 specific genomic library. The positive clone isolated, designated pKV13, was only related to F13B at the oligonucleotide region, but has proved to be a valuable chromosome 7 marker. pKV13 maps to 7pter-q22 in hybrid cell lines, and is present in a chromosome-mediated gene transfer (CMGT) cell line that also contains met and other 7q probes. pKV13 defines a common MspI restriction fragment length polymorphism (RFLP), and is genetically linked to two markers on the long arm of chromosome 7, B79a and COL1A2, both themselves linked to the cystic fibrosis locus. Multipoint linkage analysis demonstrates that KV13 maps centromeric to both B79a and COLIA2. pKV13 has been used to demonstrate the existence of rearrangements within CMGT hybrisd, and will also prove valuable in multipoint linkage studies of other 7q markers. Finally, pKV13 provides a new polymorphic locus for the characterisation of 7q deletions in myeloid disorders such as myelodysplastic syndrome.     

Novel DNA rearrangements are associated with dihydrofolate reductase gene amplification

We have employed the technique of chromosome "walking" to determine the structure of 240 kilobases of amplified DNA surrounding the dihydrofolate reductase gene in methotrexate-resistant mouse cell lines. Within this region, we have found numerous DNA rearrangements which occurred during the amplification process. DNA subclones from regions flanking the dihydrofolate reductase gene were also utilized as hybridization probes in other cell lines. Our results show that: 1) amplification-specific DNA rearrangements or junctions are unique to each cell line; 2) within a given cell line, multiple amplification-specific DNA sequence rearrangements are found; 3) the degree of amplification of sequences flanking the dihydrofolate reductase gene shows quantitative variation among and within cell lines; and 4) both the arrangement of amplified sequences as well as the magnitude of gene amplification may vary with prolonged culture even under maintenance selection conditions. These studies indicate that there is no static repetitive unit amplified in these cells. Rather, a dynamic and complex arrangement of the amplified sequences exists which is continually changing.     

Characterization of N-myc amplification in a human neuroblastoma cell line by clones isolated following the phenol emulsion reassociation technique and by hexagonal field gel electrophoresis

The N-myc amplification of human neuroblastomas was characterized by the amplified DNA cloned from the cell line MC-NB-1 using the phenol emulsion reassociation technique (PERT). A number of PERT clones exhibiting amplification in this cell line were tested for amplification in other neuroblastoma cell lines. In almost all cell lines examined, only a few clones were co-amplified with N-myc and most of the others were exclusively amplified in a subset of the cell lines. The total aggregate size of theHind III fragment identified by the PERT clones was approximately 350 kb. Most of the PERT clones were mapped to human chromosome (chr) 2p23-2pter, where the N-myc gene is located. Four types of amplicons, the 100, 420, 480 and 520 kb fragments, shown to beNot I fragments, were identified by hexagonal field gel electrophoresis. Three fragments are ordered in a head-to-tail array, and the remaining fragment is either ordered in a tail-to-head array or something else. Despite the extremely unusual construction of the amplified sequences in this cell line as compared with others, there was a low degree of sequence heterogeneity among the amplicons within this cell line. These observations lead to the idea that the complex rearrangements that give rise to the heterogeneous organization of the amplified sequences among the different cell lines precede the amplification of these sequences.     

The Gene Encoding Protein Kinase SEK1 Maps to Mouse Chromosome 11 and Human Chromosome 17

We report the mapping of the human and mouse genes encoding SEK1 (SAPK/ERK kinase-1), a newly identified protein kinase that is a potent physiological activator of the stress-activated protein kinases. The human SERK1 gene was assigned to human chromosome 17 using genomic DNAs from human–rodent somatic cell hybrid lines. A specific human PCR product was observed solely in the somatic cell line containing human chromosome 17. The mouseSerk1gene was mapped to chromosome 11, closely linked toD11Mit4,using genomic DNAs from a (C57BL/6J ×Mus spretus)F₁×M. spretusbackcross.     

Macro- and microcolinearity between the genomic region of wheat chromosome 5B containing the Tsn1 gene and the rice genome

The Tsn1 gene in wheat confers sensitivity to a proteinaceous host-selective toxin (Ptr ToxA) produced by the tan spot fungus (Pyrenophora tritici-repentis) and lies within a gene-rich region of chromosome 5B. To use the rice genome sequence information for the map-based cloning of Tsn1, colinearity between the wheat genomic region containing Tsn1 and the rice genome was determined at the macro- and microlevels. Macrocolinearity was determined by testing 28 expressed sequence markers (ESMs) spanning a 25.5-cM segment and encompassing Tsn1 for similarity to rice sequences. Twelve ESMs had no similarity to rice sequences, and 16 had similarity to sequences on seven different rice chromosomes. Segments of colinearity with rice chromosomes 3 and 9 were identified, but frequent rearrangements and disruptions occurred. Microcolinearity was determined by testing the sequences of 26 putative genes identified from BAC contigs of 205 and 548 kb in length and flanking Tsn1 for similarity to rice genomic sequences. Fourteen of the predicted genes detected orthologous sequences on six different rice chromosomes, whereas the remaining 12 had no similarity with rice sequences. Four genes were colinear on rice chromosome 9, but multiple disruptions, rearrangements, and duplications were observed in wheat relative to rice. The data reported provide a detailed analysis of a region of wheat chromosome 5B that is highly rearranged relative to rice.     

Regional localization of th human EGF-like growth factor CRIPTO gene (TDGF-1) to chromosome 3p21

The CRIPTO gene encodes a novel human growth factor structurally related to epidermal growth factor. We localized the CRIPTO gene to chromosome 3p21 by fluorescence in situ hybridization with a cosmid clone containing 40 kb of the CRIPTO genomic region (TDGF-1). To suppress hybridization to CRIPTO-related sequences, present in multiple copies in the human genome, hybridization was carried out in the presence of unlabeled CRIPTO cDNA in excess over the probe. Our finding confirms the provisional mapping of the CRIPTO gene to chromosome 3, and assigns it precisely to a chromosomal region involved in several rearrangements occurring in malignancy.CRIPTO-specific sequences are present in multiple copies in the human genome (Dono et al. 1991). Two genomic CRIPTO-encoding sequences, TDGF-1 and TDGF-3, have been isolated and characterized. TDGF-1 corresponds to the structural gene encoding the protein expressed in teratocarcinoma cells (Ciccodicola et al. 1989). TDGF-3, possibly a functional pseudogene, corresponds to a complete copy of the TDGF-1 mRNA that contains seven base changes representing both silent and replacement substitutions in the coding region (Dono et al. 1991). By somatic cell hybrid analysis TDGF-1 has been assigned to chromosome 3, and TDGF-3 to the Xq21–22 region (Dono et al. 1991).     

Loss of heterozygosity in hypotriploid cell cultures from testicular tumours

Summary We have established cell lines with a hypotriploid chromosome number from four testicular tumours. Each line had at least one Y chromosome and most of the informative centromere and enzyme markers were heterozygous implying that the tumours originated from germ cells before the first meiotic division. The small metacentric marker chromosome (i12p), specific for testicular tumours, was present in all tumour cell lines and up to three copies were found in some lines. Rearrangements of chromosome 1 and 11 were each found in three out of four tumours. The rearrangements of chromosome 1 all resulted in duplication of 1q and deletion of short-arm material from the same chromosome giving loss of heterozygosity for enzyme markers on 1p. Loss of satellite material from chromosome 13 and the centromere region of chromosome 9 were found in single cases. This study shows that even where the chromosome number of tumour cells is near triploid, regions of the genome can be deleted. The chromosomes most frequently involved in rearrangements, 1, 11, and 12 all contain sites of ras oncogenes and it is suggested that loss of normal alleles could result in homozygosity for mutant oncogenes which may play a part in tumour progression.     

Characterization of deletions in common wheat induced by an Aegilops cylindrica chromosome: detection of multiple chromosome rearrangements

An Aegilops cylindrica chromosome induces terminal deletions of chromosomes in wheat as identified by C-banding. We are constructing high-density physical maps of wheat chromosomes and have detected additional chromosome rearrangements. Among 63 lines with chromosomal subarm deletions in group 7 chromosomes, 7 lines (11.1%) were shown to harbor additional chromosome rearrangements. Two other lines were also omitted from the physical mapping because of the nature of the breakpoint calculations. The presence or absence of chromosome-specific restriction fragment length polymorphism (RFLP) or random amplified polymorphic DNA (RAPD) markers indicated that additional interstitial deletions are present in 3 lines (4.8%) with deletions in the short chromosome arms and in 4 lines (6.3%) with deletions in the long chromosome arms. We also used chromosome pairing analysis of F₁ plants of deletion lines with double ditelosomic lines of Chinese Spring wheat to detect small terminal deletions. The deletion of the most distal 1% of chromosome arm 7AL was associated with a pairing reduction of 60%.     

Development of commercially valuable traits in hexaploid triticale lines with <Emphasis Type="Italic">Aegilops</Emphasis> introgressions as dependent on the genome composition

Introgressive hybridization is an efficient means to improve the genetic diversity of cultivated cereals, including triticale. To identify the triticale lines with Aegilops introgressions, genotyping was carried out with ten lines obtained by crossing hexaploid triticale with genome-substitution forms of the common wheat cultivar Aurora: Aurolata (AABBUU), Aurodes (AABBSS), and Aurotika (AABBTT). The genome composition of the triticale lines was studied by in situ hybridization, and recombination events involving Aegilops and/or common wheat chromosomes were assumed for nine out of the ten lines. Translocations involving rye chromosomes were not observed. Substitutions for rye chromosomes were detected in two lines resulting from crosses with Aurolata. Genomic in situ hybridization (GISH) with Ae. umbellulata DNA and molecular genetic analysis showed that chromosome 1R was substituted with Ae. umbellulata chromosome 1U in one of the lines and that 2R(2U) substitution took place in the other line. Fluorescence in situ hybridization (FISH) with the Spelt1 and pSc119.2 probes revealed a translocation from Ae. speltoides to the long arm of chromosome 1B in one of the two lines resulting from crosses with Aurodes and a translocation in the long arm of chromosome 7B in the other line. In addition, the pSc119.2 probe revealed chromosome 1B rearrangements in four lines resulting from crosses with Aurolata and in a line resulting from crosses with Aurotika. The lines were tested for main productivity parameters. A negative effect on all productivity parameters was demonstrated for Ae. umbellulata chromosome 2U. The overwinter survival in all of the lines was similar to or even higher than in the original triticale cultivars. A substantial increase in winter resistance as compared with the parental cultivar was observed for the line carrying the T7BS-7SL translocation. The line with the 1R(1U) chromosome substitution seemed promising for the baking properties of triticale.