| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1
Unidad de Investigación and
2
Sección de Bioquímica, Hospital General Universitario de Guadalajara, 19002 Guadalajara, Spain;
a author for correspondence: Hospital General Universitario de Guadalajara, Unidad de Investigación, C/Donantes de Sangre, sn., 19002 Guadalajara, Spain, fax 34-49-209216, e-mail guimail{at}redestb.es
Identification of DNA mutations in urine sediments has been proposed as a noninvasive and early indicator of urinary tract cancer (1)(2)(3)(4). The sensitivity of this approach is limited by the proportion of DNA containing the mutant allele, because the resolution of the technique used to detect oncogenes or suppressor gene mutations (mainly single-strand conformational polymorphism) requires ~10% representation of mutant product (4)(5), whereas estimates to the proportion of tumor cells in the urine of bladder cancer patients were not >7% (1).
H-ras represents a single member of a family of genes coding for a 21-kDa protein involved in the regulation of cell growth and differentiation (6). Most activation events of this protooncogene to the oncogenic form found in human bladder cancer occur via a point mutation at codon 12 (7)(8). To provide a more sensitive means for detecting mutations, we have devised a sensitive, nonradioactive procedure based on two consecutive PCRs with intermediate restriction digestion, which leads to effective enrichment of mutant alleles for further amplification and elimination of wild-type alleles by digestion (5)(9)(10)(11).
Wild-type DNA was prepared from the urine sediments of three healthy volunteers, and two DNA samples containing known mutations in H-ras codon 12 were mixed with an equal quantity (200 ng) of that wild-type DNA. These mutant-type samples were extracted from tumor specimens after surgical resection. The mutations were known through direct sequencing.
Amplifications in the first round were carried out in a final volume of 100 µL in 1x PCR buffer containing 2 mmol/L MgCl2, 50 pmol of each primer, dNTPs at 40 µmol/L each, and 2.5 units of UlTma® DNA Polymerase (Perkin–Elmer–Cetus). After an initial denaturation step at 94 °C for 2 min, samples were subjected to 10 cycles of 30 s at 94 °C, 30 s at 60 °C, and 30 s at 72 °C, which were followed by 12 more cycles in which 20 s was added to the extension time after each consecutive cycle, with a final extension at 72 °C for 7 min. Primers were as follows: F (5'-GACGGAATATAAGCTGGTGGTGG-3') and R (5'-TGGATGGTCAGCGCACTCTT-3'). The 308-bp fragment amplified with these primers contains two naturally occurring MspI sites located at codon 12, which are destroyed by mutations in either of the first two positions and 55 bp upstream of codon 12, which provided a positive control for MspI cleavage.
Intermediate restriction digestion was carried out with 8 µL of PCR product with 8 U of MspI (Fermentas AB) at 37 °C overnight, giving rise to fragments of 236, 55, and 17 bp generated from wild-type PCR products and fragments of 291 bp and 17 bp from mutant products.
One microliter of undiluted restriction digest was amplified in the second PCR in a 20-µL volume in 1x PCR buffer containing 1.5 mmol/L MgCl2, 10 pmol of each primer, dNTPs at 100 µmol/L each, and 0.8 units of AmpliTaq Gold(TM) (Perkin–Elmer–Cetus). After an initial denaturation step at 94 °C for 12 min to activate the polymerase, samples were subjected to 10 cycles of 30 s at 94 °C, 30 s at 60 °C, and 30 s at 72 °C, 9 cycles in which 10 s was added to the extension time after each consecutive cycle, and 17 more cycles in which 20 s was added to the extension time after each consecutive cycle, with a final extension for 7 min at 72 °C. The following primer pairs were used: F (5'-TGAGGAGCGATGACGGAATAT-3') and R (5'-CAGGCTCACCTCTATAGTGGGGTC-3'). When this second PCR was performed, only 291-bp fragments amplify, giving rise to a 129-bp product that cleaves with MspI at one site if the codon 12 sequence is wild-type but fails to cleave if there is a mutation in the first two bases of codon 12. All PCRs were carried out in a GeneAmp® PCR System 2400 (Perkin–Elmer–Cetus), including a "no DNA" control.
The final restriction digestion was carried out overnight with 16 µL of PCR product under the conditions described above, and 20 µL of each reaction were separated on 3% Resophor® agarose gels (Eurobio), containing 0.25 mg/L ethidium bromide, at 4 V/cm in 1x Tris-acetate/EDTA buffer. A 408-bp fragment containing two cut sites was included to control for MspI digestion after the second PCR.
Mutant uncut fragments were removed from the gel, purified with the Qiaquick® gel extraction kit (Qiagen Inc.), and 2 µL were reamplified in a 100-µL volume in 1x PCR buffer containing 1.5 mmol/L MgCl2, 105 pmol of forward primer, 97 pmol of reverse primer, dNTPs at 100 µmol/L each, and 2.5 units of AmpliTaq Gold under the following conditions: 95 °C for 12 min, then 10 cycles of 95 °C for 30 s, 58 °C for 30 s, and 72 °C for 30 s, 26 more cycles in which 10 s was added to the extension time after each consecutive cycle, and a final extension at 72 °C for 7 min. The primers used were T7 forward tagged primer F (5'-GTAATACGACTCACTATAGGGCGACGGAATATAAGCTGGTGG-3') and M13 reverse tagged primer R (5'-GGAAACAGCTATGACCATGTGGATGGTCAGCGCACTCTT-3'). PCR products were purified with the Qiaquick PCR purification kit (Qiagen), and subsequent sequence analysis of these PCR fragments with both sense and antisense strands was performed using T7 and M13 reverse primers, respectively, on an ABI 373 fluorescent automated DNA sequencer (PE-Applied Biosystems).
Dilution experiments with various ratios of purified DNA containing a known wild-type or mutant H-ras gene were conducted to determine the minimal proportion of mutant DNA necessary for obtaining visible uncut bands.
To test the suitability of the assay for analyzing clinical samples, DNA was extracted from seven urine sediments, the cytology of which had been previously diagnosed as "dysplasia" and which were literally considered by the pathologists of our hospital as "suspected of having urothelial carcinoma".
The results of mutant-enriched PCR analyses of H-ras codon
12 are shown in Fig. 1
, which shows that all of the 129-bp products obtained from
wild-type samples and the 408-bp fragment used as control for the
enzyme digestion were cut again with MspI at codon 12,
yielding an 88- and a 41-bp band, and that there is no visible mutant
band. This 129-bp product arises from the amplification of uncut
wild-type fragments that resisted digestion by MspI
(5). However, the DNA mutant at codon 12 gives rise to
products that fail to cut, because of the selection of the fragments
containing the mutation in the second PCR. Sequence analysis of 129-bp
bands confirmed the presence of the previously known mutations: GGC to
CGC (Gly to Arg) and GGC to GAC (Gly to Asp).
|
A titration experiment showed that this technique is able to detect one mutant DNA copy in the presence of 900 wild-type copies. This represents the ability to detect one cell heterozygous for H-ras codon 12 in the presence of over 450 wild-type cells.
Three of the seven urine sediments tested corresponded to patients who later were found to have bladder cancer, confirmed by cystoscopy and biopsy. In one sediment, it was possible to recognize a point mutation at codon 12 (GGC to CGC; Gly to Arg). The sample was collected 38 days before diagnosis. None of the remaining samples showed visible uncut fragments.
Because the sensitivity of methods that rely on the discrimination of amplification products from point-mutated genes in a great excess of wild-type genes is limited by the error frequency of the polymerase used for the amplification, we adopted two strategies to avoid those restriction sites of wild-type fragments that may be destroyed by erroneous replacement of critical nucleotides. First, UlTma DNA Polymerase (9), which possesses 3' to 5'exonuclease proofreading activity, was used instead of Polymerase in the first amplification, and MgCl2 and dNTPs concentrations were kept low to improve fidelity. Second, cycles in the first and second PCR were limited to the critical number necessary for obtaining the amplification of uncut wild-type fragments, which were used as an internal control for efficient amplification of each sample. Moreover, the complete procedure was repeated for all samples, and each mutant sample contained the same mutation found in the first experiment. No evidence of mutation was found in the wild-type samples.
Conditions for MspI digestion are important to produce appropriate cutting of fragments: Incomplete MspI digestion means that a false-positive result could be obtained if enough of the second-round PCR product is incompletely digested. Therefore, confirmation of mutation by sequencing is unavoidable. This problem could also decrease sensitivity of the assay by increasing the usual background signal.
Mutant-enriched PCR, characterized by good practicability and high sensitivity, has previously been applied to the detection of small numbers of mutant cells against a large background of wild-type cells (5)(10)(11)(12). In the original reports, artificial, primer-mediated restriction sites were introduce into wild-type DNA by application of mismatched primers localized in the direct vicinity of possible sites of mutations. However, mutant-enriched PCR is also feasible when, as in this case, natural restriction sites are present (13).
Smearing of the staining by ethidium bromide was observed in all the lanes. That could have been caused by overloading the DNA sample (remaining primers and dNTPs) (14), because we did not dilute the product obtained with the first PCR reaction, which was used directly as the template in the second PCR reaction. Although this has a slight effect on the resolution of the bands, it eliminates the need to dilute, making the assay easier and the possibility of contamination more difficult.
The previously reported frequency of mutations in H-ras codon 12 ranges from 3% to 76%, although most authors have found a frequency around 20% (4)(7)(8)(15)(16)(17). Therefore, we cannot expect to detect all the tumors that are not clearly diagnosed through urine cytology. However, the method described above might help to reduce the length of time from initial cytology to definitive diagnosis in a percentage of patients, with a relatively low cost. Although new prospective studies must assess the diagnostic accuracy of this technique, we have demonstrated that it permits detection of mutations in clinical samples. In practice, no false positives have been seen with this assay on the four samples corresponding to patients without cancer tested, which suggests that the level of sensitivity is still insufficient to detect some background mutations in healthy individuals.
In summary, the data presented here clearly demonstrate that this mutant-enriched PCR can be applied to the detection of mutated tumor cells with excellent reproducibility. Because of the simplicity of the technique, it may suitable for screening urine sediments for mutations of the H-ras gene.
Acknowledgments
This research was supported by a grant from the Fondo de Investigación Sanitaria (96/1151) and by a grant from the Junta de Comunidades de Castilla-La Mancha (95188). José Ramón Conejo was supported by a fellowship from Fondo de Investigación Sanitaria (97/5315). We thank the Pathology Department for providing cancer samples and clinical data.
Footnotes
1 J.R.C. and T.P. contributed equally to this work 
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  S. Gelmini, A. Crisci, B. Salvadori, M. Pazzagli, C. Selli, and C. Orlando Comparison of Telomerase Activity in Bladder Carcinoma and Exfoliated Cells Collected in Urine and Bladder Washings, Using a Quantitative Assay Clin. Cancer Res., July 1, 2000; 6(7): 2771 - 2776. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
