| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Clinic of Urology and Pediatric Urology, University of the Saarland, Homburg/Saar, Germany; Departments of2 Urology and 3 Pathology, Heinrich-Heine University, Düsseldorf, Germany
aaddress correspondence to this author at: Clinic of Urology and Pediatric Urology, University of the Saarland, 66421 Homburg/Saar, Germany; fax 49-6841-1624795, e-mail urbwul{at}uniklinik-saarland.de)
Gene amplifications and deletions play an important role in the pathogenesis of solid tumors, including prostate cancer. Real-time PCR is a powerful tool for quantitative DNA analysis, particularly when starting quantities of tumor tissue are minimal (1)(2)(3)(4)(5). In the present study, we describe a modification of the 2–
CT method, which recently was shown to be suitable for relative gene expression analyses (6). We used our technique to analyze prostate cancer cell lines and tissue samples to determine the relationship between homeodomain-containing transcription factor 3.1 (NKX3.1) and MYC gene copy number alterations in this tumor type.
The specimens analyzed included (a) blood samples from healthy donors; (b) the prostate cancer cell lines DU145 (ATCC no. HTB-81) and PC3 (ATCC no. CRL-1435) and their derived sublines DU145MN1, PC3-N, and PC3-125-1L (7); (c) the colorectal cell line COLO320DM (ATCC no. CCL-220), which harbors a high-level MYC amplification; and (d) primary prostate adenocarcinoma samples obtained after radical prostatectomy from previously untreated patients. The specimens were histologically verified, and samples were taken as described previously (8). We extracted DNA from the snap-frozen prostate carcinoma samples, blood samples, and cell lines, using the Blood and Cell Culture DNA Midi Kit (Qiagen).
We performed quantitative real-time PCR using the LightCyclerTM system (Roche Diagnostics) with the FastStart DNA Master SYBR Green I LightCycler Kit (Roche Diagnostics). PCRs were run in duplicate with 20-µL reaction volumes containing 1x SYBR Green I PCR Buffer Mix, 5 mM MgCl2, 0.5 µM each primer, 1x FastStart Taq DNA Polymerase (Roche Diagnostics), and 50 ng of genomic DNA. The cycling conditions are given in Table 1
. The hot start PCR method was applied to prevent incomplete DNA denaturation as discussed by Wilhelm et al. (9). Melting curve analysis was performed for each reaction to exclude nonspecific PCR side products. We analyzed the run data, using the Second-Derivative Maximum Method (LightCycler quantification software 3.5; Roche Diagnostics). The crossing point (CP) values, defined as the points at which the fluorescence increased appreciably above background fluorescence, were averaged for each set of two reactions. For each individual run, three dilutions of COLO320DM DNA were used to generate a gene-specific internal calibration curve. Before analysis of the prostate cancer samples, PCR amplification efficiencies (E) were determined according to the equation: E = 10(–1/slope) (10). The calibration curves were created by use of five to eight dilutions containing 0.39–100 ng of COLO320DM DNA as input with three PCR repeats each (Fig. 1
, A–C). The efficiencies were as follows: 1.97 for MYC, 1.99 for NKX3.1, and 1.99 for GAPDH, each with a high linearity [Pearson correlation coefficient (r) = –1.00].
|
|
The primers used are listed in Table 1
. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH), which is located at 12p13, served as reference gene. The region 12p13 has been shown to not be involved in prostate carcinogenesis. The existence of specific primer binding sites in pseudogenes of GAPDH was excluded by use of the BLASTN database.
As a modification of the 2–CT method, we calculated the relative gene copy number from the realtime PCR efficiencies, which were determined for each individual run, and the CP deviations of the target and reference genes in a test sample vs a control: E–[(CPtarget – CPref)control – (CPtarget – CPref)test]. We determined E for each individual run by constructing an internal gene-specific calibration curve with COLO320DM DNA. Because the gene copy number is 2 in normal diploid DNA, the relative copy number multiplied by 2 yields the copy number in the test sample for the target gene.
Concerning the accuracy of our modified real-time PCR method, measurements of the MYC gene copy number in the prostate cancer cell lines revealed a high concordance with the previously reported fluorescence in situ hybridization results (7). We also found the method to be sufficiently sensitive to distinguish one copy from two copies. For this purpose, we determined the difference in copy numbers of the androgen receptor (AR) gene at chromosome Xq12 in healthy human males vs females (Fig. 1
, D and E). Our analyses showed mean (SD) relative copy numbers of 1.18 (0.18) for male DNA and 2.27 (0.34) for female DNA. We observed no difference between males and females for the autosomal gene topoisomerase II isozyme ß (TOP2B), located at 3p.
We determined the intra- and interassay variances of LightCycler runs by use of the COLO320DM cell line and DNA extracted from blood samples from healthy individuals. We investigated the intraassay variance for the MYC gene in five repeats within one LightCycler run. We calculated interassay variance from three experimental runs performed on 3 successive days with three different master premixes. The mean variance was acceptable: 0.48 cycles in intertest experiments and 0.12 cycles in intratest experiments.
To demonstrate the utility of our modified quantitative real-time PCR method, we analyzed the NKX3.1 and MYC gene copy numbers in 35 primary prostate adenocarcinomas. The detailed findings are presented in Table 1
of the Data Supplement that accompanies the online version of this Technical Brief athttp://www.clinchem.org/content/vol51/issue3/. NKX3.1 and MYC were selected because they are supposed to play an important role in prostate carcinogenesis [for a review, see Ref. (11)]. The NKX3.1 gene maps to 8p21. Authors of studies based on comparative genomic hybridization (CGH) and microsatellite/loss of heterozygosity analyses agree that 8p losses are one of the most common genetic changes in prostate cancer, with reported frequencies of 30–86% in different studies (12). Amplification and overexpression of the MYC protooncogene at 8q24 have been observed in 8% of primary prostate cancers, increasing to 
30% in advanced/metastatic tumor stages (13)(14).
To determine whether results obtained by real-time PCR analysis of the prostate cancer samples were significantly different from the mean results obtained for samples from healthy individuals, we determined a tolerance interval (TI) for the relative gene copy numbers, using the mean (SD) CP values for target and reference genes in 11 healthy individuals according to the equation: TI = 2 ± (SDCP x 2). The TI ranged from 1.44 to 2.56 for NKX3.1 and from 1.63 to 2.37 for MYC. To distinguish chromosome 8 aneusomy from gene-specific changes, we determined the TI for the copy number ratio of NKX3.1 to MYC; the resulting values ranged from 0.77 to 1.23. We compared the PCR results for the tumor tissue samples with the CGH findings on aliquots of the same DNA samples. CGH was performed as described elsewhere (7).
CGH detected loss of genetic material at 8p in 6 of 35 (17.1%) cases analyzed. This relatively low incidence of detected 8p losses contrasts to the 18 cases (51.4%) for which real-time PCR revealed a significant decrease in NKX3.1 gene copy number, from 1.39 to 0.20 copies/cell. Conversely, in three of the six cases with CGH-detected loss at 8p, real-time PCR detected no changes in the NKX3.1 gene copy number. CGH detected a gain in genetic material at 8q in 9 cases (25.7%), whereas PCR detected an increase in MYC in 13 cases (37.1%; from 2.45 to 8.46 copies/cell). Only three the cases for which CGH detected a gain at 8q had a MYC gain according to PCR. PCR detected a decrease in MYC gene copy number in six tumors, one of which also showed loss of genetic material at 8q24 by CGH.
The ratio of NKX3.1 to MYC was below the TI in 22 cases (63%), comprising distinct classes of genetic alterations: loss of NKX3.1 and normal MYC (8 tumors), normal NKX3.1 and gain of MYC (4 tumors), or loss of NKX3.1 concurrent with gain of MYC (7 tumors). CGH did not detect concurrent 8p loss and 8q gain in any of the seven tumors. Conversely, two tumors with concurrent loss at 8p and gain at 8q according to CGH were not detected by PCR. Four tumors exhibited loss of both genes or gain of both genes, leading to a significant decrease in the ratio of NKX3.1 to MYC below normal in three cases. Our findings concur with recent observations of an association between loss of genetic material at 8p and gain at 8q in prostate cancer (15)(16).
In summary, our relative real-time PCR quantification method based on the LightCycler system differs from previous methods in that it uses the fluorescent dye SYBR Green I, which rapidly and sensitively detects low amounts of DNA molecules. Compared with hybridization probes, SYBR Green I is less expensive, which becomes particularly important when screening of larger series of tissue samples is intended. As a modification of the 2–CT method, we generated internal calibration curves to determine the gene-specific PCR efficiencies for each individual run. This approach allows for screening analysis of individual genes in separate runs under optimum PCR conditions. As demonstrated for prostate cancer samples, our real-time PCR quantification method is suitable for screening of specific genes and could be suitable for other applications in which gene copy number alterations are an important feature of pathogenesis.
Acknowledgments
We greatly appreciate the excellent technical assistance of Christiane Hader. The study was supported by the Deutsche Krebshilfe (70-2936-Wu I).
References
CT method. Methods 2001;5:402-408.
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  C. O'Brien, G. Cavet, A. Pandita, X. Hu, L. Haydu, S. Mohan, K. Toy, C. S. Rivers, Z. Modrusan, L. C. Amler, et al. Functional Genomics Identifies ABCC3 as a Mediator of Taxane Resistance in HER2-Amplified Breast Cancer Cancer Res., July 1, 2008; 68(13): 5380 - 5389. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Jung, R. Kindich, J. Kamradt, M. Jung, M. Muller, W. A. Schulz, R. Engers, G. Unteregger, M. Stockle, R. Zimmermann, et al. Genomic and Expression Analysis of the 3q25-q26 Amplification Unit Reveals TLOC1/SEC62 as a Probable Target Gene in Prostate Cancer Mol. Cancer Res., March 1, 2006; 4(3): 169 - 176. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
