Authentication of nasopharyngeal carcinoma tumor lines

Abstract

Nasopharyngeal carcinoma (NPC) is rare in most part of the world but shows high incidence in South China and Southeast Asia. One of the unique features of this epithelial cancer is its consistent association with Epstein–Barr virus (EBV) infection. Studies have shown that EBV plays a key role in the genesis and maintenance of the tumor phenotype in this disease.1 Clonal EBV genome and latent gene products are found in almost all cases of NPC expect for the keratinizing type of NPC from nonendemic areas. With the advance of molecular biology, knowledge of molecular alterations and EBV's function in NPC rapidly accumulated in the past 15 years.1, 2 However, most of these studies are based on the EBV-negative NPC cell lines established at 10–20 years ago. Until now, only a few NPC cell lines, such as C666-1, stably harboring the EBV genome were used as EBV-positive NPC model.3 In addition to the established NPC cell lines, several EBV-positive NPC transplants in nude mice (e.g. C15, C17, xeno-666, xeno-2117 and xeno-1915) were used for identifying the genetic alterations and delineating aberrant signal transduction pathways in NPC.4, 5 Because of the difficulties in manipulating target gene expression in these in vivo models, they are rarely used in functional studies. Recently, we have performed the array-based CGH analysis on a number of NPC cell lines, xenografts and primary tumors.6 We observed similar copy number aberrant patterns in some EBV-negative cell lines. Furthermore, these cell lines show distinct transcription profiles from the C666-1 cell line and NPC xenografts in microarray analysis (unpublished observations). It raised the concern for the identity of tumor lines used in NPC study. Thus, we investigated the authenticity of a number of NPC cell lines (C666-1, HK-1, CNE-1 and CNE-2), xenografts (xeno-666, xeno-2117, xeno-1915, C15 and C17) and immortalized nasopharyngeal epithelial cell lines (NP69 and NP460), which are commonly included in our previous works.1 The DNA samples from these tumors were extracted and subjected to DNA fingerprinting analysis using the AmpF/STR Identifiler® PCR Amplification Kit (Applied Biosystems, Foster City, USA). A total of 15 short tandem repeat (STR) loci (D3S1358, vWA, FGA, D8S1179, D21S11, D18S51, D5S818, D13S317, D7S820, D16S539, TH01, TPOX, CSF1P0, D2S1338, D19S433) were co-amplified in each sample and detected on an ABI 3100 Genetic Analyzer according to the manufacturer's protocol. Amelogenin loci at the sex chromosomes was also examined. The data were analyzed and allele(s) of each locus were determined by GeneScan and GeneMapper™ ID Software (Applied Biosystems, Foster City, USA). The STR profiles of NPC tumors and immortalized nasopharyngeal epithelial cell lines are shown in Table I. Distinct STR profiles were found in HK-1, NP69, NP460 and all EBV-positive xenografts (xeno-666, xeno-2117, xeno-1915, C15 and C17). The STR profile of xeno-666 and its devised cell line C666-1 are almost identical.3 In C666-1 cells, multiple alleles of D8S1179 and D21S11 loci were observed. It may be due to the microsatellite instability feature of this cell line. Surprisingly, we found high similarity between the STR profiles of 2 EBV-negative cell lines, CNE-1 and CNE-2. Complete identical alleles were detected in 12 out of 15 loci. Compared to CNE-1, gain of an extra allele at TPOX and loss of 1 allele at FGA and THO1 was observed in CNE-2. The minor allelic variations may be due to genetic instability or chromosomal aberrant during long-term in vivo culture. The finding implies that these 2 widely used EBV-negative cell lines are genetically identical and should be derived from a single cell line. Because many studies have shown that a number of cell lines were contaminated with HeLa,7 we then investigated whether CNE-1 and CNE-2 are contaminated with this common cell line. HeLa is a cell line devised from the glandular cancer of cervix, which contains the HPV-18 genome. Nevertheless, HPVs have not been detected in primary NPC tumors. We have examined the presence of E6/E7 sequence of HPV-18 in all NPC samples by PCR analysis. Interestingly, HPV-18 sequences were only found in the CNE-1 and CNE-2 cells (Fig. 1). Furthermore, we examined whether the STR profile of the HeLa is identical to that of the 2 EBV-negative NPC cell lines. The STR profile of HeLa determined was completely matched with that listed in the ATCC datasheet. While we compared the STR profiles among HeLa, CNE-1 and CNE-2, we noted that the CNE-1 and CNE-2 profiles contain at least 1 identical allele from HeLa for the 15 loci examined. Interestingly, in several loci (e.g. vWA, D8S1179, D7S820, CSF190, D19S433), only 1 of 2 alleles from HeLa was detected in the CNE-1 and CNE-2 genomes. The presence of partial genome of HeLa in CNE-1 and CNE-2 excludes the possibility that the cell lines are simply cross-contaminants of HeLa cells. Our previous cytogenetic study also did not found 2 heterogenous popula