| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
Departments of1 Chemical Pathology, 2 Clinical Oncology, 3 Anatomical and Cellular Pathology, and 4 Medicine and Therapeutics, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, New Territories, Hong Kong SAR, China;
aaddress correspondence to this author at: Department of Chemical Pathology, Room 38061, 1/F, Clinical Sciences Bldg., Prince of Wales Hospital, 30-32 Ngan Shing St., Shatin, New Territories, Hong Kong SAR; fax 852-2194-6171, e-mail loym{at}cuhk.edu.hk
Despite the increasing number of clinical applications of circulating Epstein–Barr virus (EBV) DNA analysis for the detection (1)(2), monitoring(3)(4), and prognostication(5)(6) of nasopharyngeal carcinoma (NPC), several fundamental questions concerning plasma EBV DNA, including its tissue origin and relationship with tumor mass, remain unanswered.
In the first part of this study, we investigated the relative contributions of tumor cells and other latently EBV-infected lymphoid tissues to the pool of circulating EBV DNA. Previous studies have shown that different EBV genotypes are harbored by latently EBV-infected lymphoid tissues and blood cells (7)(8)(9)(10) in healthy persons and patients with malignancies. On the other hand, it is well established that the EBV is monoclonal in NPC tumor tissues and other EBV-associated cancers(11)(12)(13)(14). Therefore, investigation of the genotype of plasma EBV DNA should reveal the relative contributions of tumor cells and other latently EBV-infected lymphoid tissues to the pool of circulating EBV DNA in NPC patients.
For this study, 25 patients with newly diagnosed NPC, 15 patients with systemic lupus erythematosus, and 13 renal transplant recipients were recruited and gave informed consent. This study was approved by the ethics committee of the Prince of Wales Hospital, Hong Kong. Plasma samples were collected from all patients, and tumor specimens for microdissection were obtained from 14 of the 25 NPC patients. DNA extracted from the plasma samples and the microdissected tumor tissues was used for the amplification of EBV DNA, targeting a region encoding the carboxy terminus of latent membrane protein (LMP-1), using the primers 5'-ATGGTAATGCCTAGAAGTAAAGAAAGG-3' (LMP1f) and 5'-CATAGCCCTAGCGACTCTGCTG-5' (R2b). The PCR products were ligated to the pGEM®-T Easy Vectors (Promega), and the ligation products were transformed into competent Escherichia coli cells. Twelve bacterial clones were picked and sequenced with a BigDye® Terminator (Ver. 1.1) Cycle Sequencing Kit (Applied Biosystems). The sequences were aligned by the SeqScape® (Ver. 2.0) software (Applied Biosystems).
EBV DNA encoding the LMP-1 was successfully amplified from the plasma of all 25 NPC patients and 8 patients receiving immunosuppressive therapies, including 3 patients with systemic lupus erythematosus and 5 renal transplant recipients. The EBV genotypes detected in the plasma of the NPC patients and the immunosuppressed patients are shown in Table 1
(also see Table 1
in the Data Supplement that accompanies the online version of this Technical Brief at http://www.clinchem.org/content/vol51/issue11). The sequences are referenced to the sequence of EBV strain B95-8 (accession no. V01555). Only the differences between the patients’ sequences and the B95-8 sequence are listed in Table 1
. A locally predominant sequence was detected in the plasma of 15 NPC patients (patients N1 to N15) and 6 immunosuppressed patients (patients S1, S2, and R1 to R4). This locally predominant sequence is characterized by a 30-bp deletion (15) compared with the B95-8 sequence. Only 1 EBV genotype was detected in the plasma of 22 (88%) of the 25 NPC patients. In the remaining 3 NPC patients (12%; patients N13 to N15), 2 EBV genotypes were detected in the plasma. The nucleotide sequences of the minor and major genotypes differed by a single nucleotide substitution in each of the 3 patients. The minor genotype represents 11%–22% of the total number of clones in these 3 patients. With regard to the representation of the number of clones to the actual composition of plasma EBV DNA, Walling et al.(8) have shown that the ratios of the resulting clone sequences in an EBV genotyping assay would closely resemble the original sequence ratios before PCR. With regard to the tissue origin of plasma EBV DNA, we have shown that in each of the 14 patients with available tumor tissues, the LMP-1 nucleotide sequence of the predominant genotype was identical to the sequence of the tumor. In 8 of these 14 patients (N18 to N25), the EBV genotypes showed characteristic nucleotide changes that distinguished them from the locally predominant genotype. Our findings suggest that most of the circulating EBV DNA molecules in NPC patients are tumor derived.
|
In contrast, only samples from 3 (37.5%) of the 8 patients receiving immunosuppressive therapies (patients S1, R1, and R2) showed a single EBV genotype. Samples from 4 of the 8 patients (patients S2, S3, R3, and R5; 50%) showed 2 EBV genotypes, and the samples from 1 patient (R4; 12.5%) showed 3 EBV genotypes. In these 5 patients, the minor and the major genotypes differed by substitutions of 1 to 2 nucleotides in the nucleotide sequence. For the 4 patients showing 2 EBV genotypes, the minor genotype comprised 29%–40% of the total number of clones. For the patient showing 3 EBV genotypes (R14), there were 2 codominant EBV genotypes, and each of them accounted for 43% of the total number of clones. The proportions of plasma samples showing a single EBV genotype were significantly higher in the NPC group than in the group receiving immunosuppressive therapies (22 of 25 vs 3 of 8; P = 0.01, Fisher exact test).
To ensure that the observed difference in the plasma EBV DNA genotype patterns between NPC patients and immunosuppressed patients was not attributable to the difference in EBV DNA template concentrations, we repeated the experiments with diluted plasma DNA samples from 5 NPC patients, using concentrations of EBV DNA comparable to those in samples from immunosuppressed patients as PCR templates. Plasma samples from all 5 patients showed a single LMP-1 nucleotide sequence in the initial analyses. In the repeat experiments, the sample from each of the 5 patients showed only 1 EBV genotype, and the LMP-1 nucleotide sequence was identical to that of the initial analysis. These differences in the genotype patterns and compositions of plasma EBV DNA in the 2 groups of patients thus suggest that the tissue origins of their plasma EBV DNA were different.
In the second part of the study, we established a nude mouse model to study the relationship between the tumor mass and plasma EBV DNA concentration. One EBV-positive xenograft (Xeno-2117) (16), 1 EBV-positive cell line (C666)(17), and 1 EBV-negative cell line (HK1)(18) were used for the study of the relationship between tumor mass and plasma EBV DNA concentration in nude mice. We injected 5 x 106 to 1 x 107 tumor cells (0.3–0.4 mL) subcutaneously into the flanks of anesthetized nude mice. The xenografts were allowed to grow for 2–9 weeks to achieve tumors of different sizes. Quantitative real-time PCR was used to measure the plasma concentrations of EBV DNA, human nuclear DNA (by targeting the albumin gene), human mitochondrial DNA [by targeting NADH dehydrogenase subunit 6 (ND6)], and murine DNA (by targeting the murine ß-globin gene). The sequences of the primers and probes and the thermal profiles of the real-time PCRs are listed in Table 2 in the online Data Supplement.
EBV DNA was detected in the plasma of all nude mice inoculated with EBV-positive cells (C666 cell line and Xeno-2117). The plots for the correlations between tumor mass and the concentrations of plasma EBV DNA and human albumin gene are shown in Fig. 1
. There was a linear relationship between the tumor mass and the plasma EBV DNA concentrations. The regression coefficients (r2) for the mice inoculated with the C666 cell line and Xeno-2117 tumor cells were 0.750 and 0.914, respectively. A linear correlation also existed between tumor mass and the plasma concentrations of human albumin gene (r2 = 0.746 and 0.625 for C666- and Xeno-2117–inoculated mice, respectively). In the 4 nude mice inoculated with HK1 cells, an EBV-negative cell line, the human albumin gene, human mitochondrial ND6 gene, and murine ß-globin gene sequences, but not EBV DNA sequences, were detected in the plasma. The plasma human albumin gene concentrations showed a linear relationship with the tumor mass (r2 = 0.808). Because the only source of EBV DNA and human DNA in the mouse model was the engrafted tumor, the plasma concentrations of EBV DNA and human albumin gene would represent the concentrations of tumor-derived nuclear DNA. Our findings therefore suggest a direct linear relationship between the plasma concentrations of tumor-derived nuclear DNA and the tumor mass.
|
On the other hand, we found no linear correlation between the tumor mass and the human mitochondrial ND6 gene (r2 = 0.123 and 0.374) or the murine ß-globin gene (r2 = 0.363 and 0.037) concentrations in mice inoculated with C666 and Xeno-2117 tumor cells. We measured the human mitochondrial ND6 gene concentrations in the resected tumor tissues and found that there was a wide variation of up to 5-fold even within the same type of cancer tissues (data not shown). One explanation would be the variable number of mitochondria in each tumor cell and the variable number of mitochondrial DNA genomes in each mitochondrion (19). In the 5 mice not inoculated with tumor cells, only the murine ß-globin gene was detected in the plasma.
In summary, our results substantiate the tumoral origin of circulating EBV DNA in NPC patients and show that the concentrations of EBV DNA and other tumor-derived nuclear DNA sequences vary in a linear fashion with the mass of the NPC.
Acknowledgments
The section of the text concerning plasma EBV genotyping was supported by a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (Project CUHK 4086/02M), and the section correlating tumor mass and plasma EBV DNA concentrations was supported by the Kadoorie Charitable Foundations (under the auspices of the Michael Kadoorie Cancer Genetics Research Program).
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  K. C. A. Chan, P. B. S. Lai, T. S. K. Mok, H. L. Y. Chan, C. Ding, S. W. Yeung, and Y. M. D. Lo Quantitative Analysis of Circulating Methylated DNA as a Biomarker for Hepatocellular Carcinoma Clin. Chem., September 1, 2008; 54(9): 1528 - 1536. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.C. A. Chan, S. F. Leung, S. W. Yeung, A. T.C. Chan, and Y.M. D. Lo Quantitative Analysis of the Transrenal Excretion of Circulating EBV DNA in Nasopharyngeal Carcinoma Patients Clin. Cancer Res., August 1, 2008; 14(15): 4809 - 4813. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K.C. A. Chan, S.-F. Leung, S.-W. Yeung, A. T.C. Chan, and Y.M. D. Lo Persistent Aberrations in Circulating DNA Integrity after Radiotherapy Are Associated with Poor Prognosis in Nasopharyngeal Carcinoma Patients Clin. Cancer Res., July 1, 2008; 14(13): 4141 - 4145. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
