| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Department of Obstetrics and Gynecology, Division of Gynecology, Medical University of Vienna, Vienna, Austria;2 Ludwig Boltzmann Institute for Gynecology and Gynecologic Oncology, Vienna, Austria;3 ViennaLab Labordiagnostika GmbH, Vienna, Austria;4 Biofocus GmbH, Recklinghausen, Germany
aaddress correspondence to this author at: Department of Obstetrics and Gynecology, Division of Gynecology, Medical University of Vienna, Vienna, Austria, Waehringer Guertel 18-20, 1090 Vienna, Austria; fax 43-1-40400-7832, e-mail robert.zeillinger{at}meduniwien.ac.at
Ovarian carcinoma is the fifth most common female cancer type and the most common cause of death from gynecologic malignancies in the Western world (1). The three members of the ras gene family, H-ras, K-ras, and N-ras, are among the most common oncogenes associated with human neoplasms (2). Mutations in the K-ras gene are frequently found in malignant neoplasms: 90% of adenocarcinomas of the pancreas; 50% of colon, 30% of lung, and 50% of thyroid tumors; and 30% of myeloid leukemia cases, respectively (3). K-ras-activating mutations occur in codons 12 and 13 and seldom in codon 61, and lead to constitutive activation of the protein by increasing GDP/GTP exchange or decreasing GTPase activity of the protein, thus leading to increased cell proliferation.
K-ras mutation frequencies seem to be highly related to tumor histology. In general, K-ras mutations occur more frequently in mucinous tumors, including borderline malignancies, than in nonmucinous tumors such as serous carcinomas (4)(5)(6)(7)(8). K-ras mutations are more common in borderline serous tumors than in serous carcinomas, suggesting distinct etiologies (5)(9)(10)(11).
A biochip application for detection of the 10 most common mutations of K-ras codons 12 and 13 (12) combines mutant-enriched amplification with a highly specific hybridization protocol. The chip appears suitable for the detection of K-ras mutations in human feces (12). An improved biochip platform, called GeneStiX (ViennaLab Labordiagnostika GmbH), is designed to meet the needs of molecular diagnostic applications. Up to 400 different DNA capture oligonucleotides can be immobilized on the tip of a special plastic stick contained in a cylindrical tube. This allows hybridization with low volumes in a closed system (tube) and the use of standard laboratory equipment, such as a thermoshaker (Fig. 1
).
|
To evaluate the compatibility of the GeneStiX system with rapid mutation screening in tumor tissue, we analyzed ovarian tumor specimens for the presence of variations in the K-ras gene. We did not study K-ras codon 61 mutations because of their reported low frequency in ovarian carcinomas (13)(14)(15).
We collected 85 ovarian tumor specimens from patients seen at the Department of Obstetrics and Gynecology at the University of Vienna School of Medicine between 1991 and 1997. The histologic types included 36 serous carcinomas, 10 mucinous carcinomas, 11 endometrioid carcinomas, 5 clear cell neoplasms, 5 undifferentiated carcinomas, 1 mixed tumor (serous undifferentiated), 6 nonepithelial tumors, and 11 tumors of borderline malignancy (5 mucinous, 5 serous, and 1 endometrioid). DNA was isolated by use of commercially available DNA extraction reagents (DNA Extraction System I; ViennaLab Labordiagnostika GmbH) and was stored at 4 °C until analyzed.
K-ras mutant-enriched PCR amplification and K-ras GeneStiX hybridization were done according to the manufacturer’s protocols. Briefly, downstream primers were biotin-labeled, and upstream primers were phosphorylated at the 5' position. A 10-µL portion of the PCR products was digested with 1 µL of 
-Exonuclease (New England BioLabs, Inc.) for 30 min at room temperature before dilution with the provided assay buffer, which included a hybridization control oligonucleotide. The exonuclease-treated PCR product was then transferred to a GeneStiX tube containing 120 µL of hybridization buffer. Hybridization of the GeneStiX was performed at 37 °C for 1 h in a conventional thermoshaker (Eppendorf AG) to ensure adequate temperature control and constant mixing. Without additional washing steps, the biochip was stained for 10 min with the provided streptavidin–horseradish peroxidase conjugate and chemiluminescence substrate. Sticks were then rinsed with 2 mL of assay buffer and analyzed with the GeneStiX-Imager, a chemiluminescence detector developed for use with the GeneStiX system (Fig. 1
). Images were automatically processed with the test-specific analysis software provided with the imager, and a report was generated for each sample.
We identified 17 samples with K-ras mutations; 15 (88%) were positive for codon 12 mutations, and 2 (12%) were positive for codon 13 K-ras mutations. Seven of 17 (41%) mutations found in our study were Asp12, followed by 3 tumors containing the Val12 mutation, 3 tumors containing the Cys12 mutation, 2 tumors containing the Asp13 mutation, 1 containing the Arg12 mutation, and 1 containing both the Ala12 and Asp12 mutations. Mutations at codon 12 of the K-ras gene were present in 27% (3 of 11) of borderline tumors, 50% (5 of 10) of mucinous ovarian carcinomas, 14% (5 of 36) of serous carcinomas, and 18% (2 of 11) of endometrioid carcinomas (Table 1
). One serous and two mucinous borderline tumors contained mutations in codon 12. Gly-to-Asp (GGT
GAT) mutations at codon 12 of the K-ras gene were common for two borderline tumors (one mucinous and one serous), two mucinous ovarian carcinomas, one clear cell neoplasm, one endometrioid tumor, and one nonepithelial tumor, but not for the serous ovarian carcinomas. Remarkably, for the five serous ovarian carcinomas positive for K-ras mutations, the codon 12 Gly-to-Cys (GGTTGT) mutation occurred exclusively in three cases.
|
K-ras mutation status was not correlated with either FIGO stage or histologic type, whereas the presence of K-ras mutations was significantly higher in well-differentiated ovarian tumors than in moderately or poorly differentiated ones (P = 0.026). This finding is consistent with recently published data and supports the hypothesis that well-differentiated ovarian tumors and moderately or poorly differentiated ones develop along different pathways (16)(17)(18).
All mutations found by GeneStiX hybridization were confirmed by sequencing after mutant-enriched PCR amplification, which is in agreement with former validation experiments (12). However, mutations were not detectable by sequencing without previous enrichment of mutants. For specificity and sensitivity, peptide nucleic acid (PNA) mediates preferential amplification of mutant K-ras sequences (19). We evaluated the sensitivity of the amplification specific to K-ras mutants by use of a dilution series of DNA mutant for K-ras (extracted from SW480 cells) mixed with wild-type DNA (Colo320 cells). Even in a 1000-fold excess of wild-type DNA, GeneStiX hybridization unambiguously identified a single mutation.
Although a capture probe for the wild type is present on the biochip, wild-type DNA is usually not detected because of the presence of excess PNA, which is complementary to the wild-type capture probe. Actually, detection of wild-type DNA is already prohibited by the PNA-clamped amplification procedure, which ideally would generate only mutated amplification products. However, in some cases, amplification of the wild-type sequence is not completely suppressed, leading to a visible band in a gel but lacking a signal for a mutation after hybridization on the biochip. As evidenced by sequencing, these amplification products consist of wild-type sequences obviously not suppressed by PNA clamping (12). Rarely, hybridization of the amplified wild-type product can lead to an increase in background hybridization signal, mainly nonspecific for Gly12 and/or Asp13. In this case, a relatively strong hybridization signal for the wild-type capture probe compared with the probes for mutant K-ras is generated, which is indicative of an unexpected wild-type amplification during PNA-PCR. Thus, the sensitivity of the introduced K-ras mutation detection system is attributable to PNA-PCR, whereas biochip hybridization is required to improve detection specificity rather than enhancing sensitivity. To detect a single mutation, several quality criteria, which have been described by Prix et al. (12), must be fulfilled and are automatically checked by the test-specific analysis software. Problems with the GeneStiX system occur if the amount of mutated DNA is >1000-fold lower than the amount of wild-type DNA. In those instances, the reproducibility of the hybridization signal intensity is reduced because of decreased sensitivity and specificity of the PNA-PCR. Consequently, hybridization efficiency is decreased for the three spotted capture probes indicative of a single K-ras mutation.
As in other studies, K-ras mutations were less common in serous ovarian tumors (15%) than in mucinous lesions (47%) (4)(6)(7)(8)(16)(17)(20) and more common in borderline tumors (27%; 3 of 11) than in invasive cancers (19%; 14 of 74) (16). Additionally, we found that 40% of the investigated mucinous ovarian tissue specimens had a mutation in K-ras codon 12, which was the highest detection frequency in a subgroup of patients in this study and significantly higher than the frequency of 12% for serous ovarian tissue specimens. These findings are in line with previously reported data (6)(8). In addition to serous borderline tumors, one mucinous borderline tumor and two mucinous ovarian carcinomas exhibited the same mutation at codon 12 (Gly to Asp). Ovarian tumors of borderline malignancy might represent a pathologic continuum between benign and invasive carcinoma. This observation was made only for the mucinous subtype, indicating that initiation and progression of serous carcinomas might be different. Interestingly, the K-ras status in this study supports the view that some serous and mucinous borderline tumors might develop along the same pathways. In that case, genes other than K-ras might be involved in the development of different histologic subtypes of borderline tumors arising from benign tissues. Alterations in the K-ras oncogene may be clinically important with respect to tumor etiology, early diagnosis, and prognosis. Whether K-ras mutation analysis will also have an impact on therapeutic intervention remains to be seen.
We conclude that the GeneStiX system is well suited for analyzing K-ras mutations in tumor tissue specimens. The protocol presented here is fast, easy to perform, and reproducible. The stick-in-a-tube principle is convenient for daily use in routine diagnostics. The reports generated by the software are sufficient for data analysis, and no additional calculations are necessary. The results of this study are consistent with those reported previously and strengthen the thesis that some mucinous borderline tumors may progress to mucinous ovarian carcinomas based on the finding of the same K-ras mutations in both ovarian tumor subtypes. For serous ovarian carcinomas, K-ras status suggests an alternative tumorigenic pathway that differs from that of mucinous ovarian carcinomas.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  Z. Z. Zhao, D. R. Nyholt, L. Le, N. G. Martin, M. R. James, S. A. Treloar, and G. W. Montgomery KRAS variation and risk of endometriosis Mol. Hum. Reprod., November 1, 2006; 12(11): 671 - 676. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
