Terry H. Landowski, Ibrahim Buyuksal, and William S. Dalton H. Lee Moffitt Cancer Center and Research Institute, University of South Florida, 12902 Magnolia Drive, Tampa FL 33612
Identifying of mutations within specific genes has become an important strategy in determining the diagnosis and prognosis for many diseases, including cancer. Several techniques have been developed to examine heterogeneous tissue samples for specific mutations, including restriction fragment length polymorphism (RFLP), single stranded conformation polymorphism (SSCP), and denaturing gradient gel electrophoresis (DGGE), among others. These techniques are useful for rapid screening of heterogeneous tissues, and each has its individual strength.1 Using RT-PCR and DGGE, we have examined the cytoplasmic region of the Fas antigen known as the death domain in tumor specimens from patients with lymphoma of various histiocytic origins. These mutations may contribute to the failure of the tumor cells to undergo apoptosis, and thus contribute to the pathogenesis and progression of the disease.2
Materials and Methods
Tumor biopsies were obtained during postoperative resection, and peripheral blood lymphocytes for controls were obtained from normal volunteers. Total RNA was isolated from cryopreserved specimens and 50-100 ng of RNA was reverse transcribed by oligo dT priming. PCR primers for the FasIII region were as previously described2 with the addition of a 42 bp GC clamp on the forward primer.3 Following 35 cycles of amplification in a 9600 Perkin Elmer thermocycler, the PCR products were denatured for 5 minutes at 95 C and reannealed for 1 hour at 55 C to allow heteroduplex formation.
Melting temperature of the 430 bp PCR product was predicted using MacMelt software, and confirmed by analysis on a 6% acrylamide gel (37.5:1) with a perpendicular gradient of 12.537.5% denaturant. Electrophoresis in the DCode system was at a constant temperature of 60 C in 1x TAE buffer. Gels were run at 100 V for 2.5 hours. Optimum strand separation occurred at 2324% denaturant, which correlated to the predicted melting temperature of 6570 C. All further analyses were done in a parallel gradient of 1530% denaturant at 60 C, with running time extended to 4.5 hours. PCR products were detected by ethidium bromide staining.
Results and Discussion
Hematological tumors typically contain a heterogeneous population of cells that includes both neoplastic and normal cells. The identification of tumor specific mutations requires a method that can separate mutant sequences from normal sequences for isolation and identification. Sensitivity of the techniques is important in diseases such as lymphoma, which typically contain large numbers of normal stromal cells and infiltrating lymphocytes.
To determine the sensitivity of the DGGE analysis, we took advantage of a previously identified polymorphism in the Fas antigen, in which bp836 is changed from C to T.4 Direct sequencing of PCR products has shown that the T cell leukemia cell line CEM is heterozygous for the alternative codon, ACT, while the myeloma cell line 8226 expresses only the higher frequency codon, ACC. RNA was isolated from CEM and 8226 cell lines, and titrated at 1%, 5%, 10%, 30%, and 50% to determine the minima l amount of polymorphic cDNA detectable (Figure 1). Using ethidium bromide staining, we were able to easily identify 5% polymorphic RNA, and even as little as 1% was detectable.
Tumor specimens from patients with lymphoma were examined for mutations in the signal transducing domain of the Fas antigen by RT-PCR and DGGE analysis. Figure 2 shows a representative ethidium stained DGGE gel including 10 patient samples. The CEM T cell line serves as a positive control, and demonstrates the heteroduplex bands formed by polymorphic and wild type cDNA. The myeloma cell line 8226, which serves as a negative control, has been fully sequenced and shown to be wild type sequence. Of the patient specimens demonstrated in Figure 2, two display a single band characteristic of wild type sequence. Patient specimens 4, 5, 7, 9, and 10 demonstrate heteroduplex band formation, indicating an alteration in the cDNA sequence. These bands were extracted from the gel and reamplified by PCR for direct sequencing. In addition, the altered mobility of the single band seen in patient 8 is suggestive of a homozygous mutation. This band was also extracted for sequence analysis.
With the high frequency of polymorphisms reported at bp 836 in the Fas antigen, we sought to develop a screening method to distinguish the known polymorphism from other unknown mutations that could have functional significance. Expression of the alternative codon, ACC, introduces a unique restriction site for the enzyme DraI which is not present in the wild type cDNA sequence. Cleavage of the FasIII cDNAs with DraI results in an 368 bp fragment when the polymorphic codon is expressed, while this enzyme does not c ut the wild type cDNA. Patient samples that demonstrated heteroduplex formation on DGGE analysis were incubated overnight at 37 C with 0.5U DraI, denatured for 5 minutes at 95 C, and allowed to reanneal at 55 C for 1 hour. Digested products were loaded on the DGGE acrylamide gel, and electrophoresed 3.5 hours under conditions described above (Figure 3). This treatment resulted in the disappearance of one of the heteroduplex bands in specimens with the polymorphism and the appearance of a truncated homoduplex band (see lanes 1, 4, 5, 8, and 9 of Figure 3). Patient specimens that are homozygous for the polymorphism display only the truncated homoduplex (lane 10), while patient specimens with mutations in addition to the polymorphism retained two heteroduplex bands (lanes 7 and 11).
Crosslinking of the Fas antigen by antibody or specific ligand induces apoptosis in susceptible cells. Mutations within the signal transducing domain may render the cells resistant to programmed cell death, and contribute to the progression of neoplastic disease. We have utilized the methods of RT-PCR and DGGE to detect small changes in the coding sequence of the Fas antigen in hematological tumors.
In contrast to many other rapid screening techniques, DGGE is based on known physical parameters of a particular gene. By calculating the predicted melting temperature of a segment of DNA, analytical conditions can be optimized for maximum accuracy and sensitivity.5 We have demonstrated that DGGE can be used to detect as little as 1% of total RNA containing a single base change with non-radioisotopic techniques. In addition, with a combinati on of restriction digest and DGGE analysis, we have developed a technique that can be used to distinguish between previously characterized non-coding changes in the Fas antigen, and unknown mutations. Identification of the known polymorphism reduces the number of bands which must be fully sequenced, and contributes to the overall efficiency of the genetic screening.
1. Eng, C. and Vijg, J. Nature Biotech., 15, 422426 (1997).
2. Landowski, T. H., Qu, N., Buyuksal, I., Painter, J. S. and Dalton, W. S., Blood (1997).
3. Sheffield, V. C., Cox, D. R., Lerman, L. S. and Myers, R. M. Proc. Natl. Acad. Sci. USA, 86, 232236 (1989).
4. Fiucci, G. and Ruberti, G. Immunogen., 39, 437439 (1995).
5. Myers, R. M., Maniatis, T. and Lerman, L. S., Meth. Enz., 155, 501527 (1987).
The Polymerase Chain Reaction (PCR) process is covered by patents owned by Hoffmann-LaRoche. Use of the PCR process requires a license.
back to top