Michael Hepburn and Glenn A. Miller, Ph.D., Molecular Profiling Laboratory, Genzyme, Framingham, MA
The search for sequence variations in genomic DNA is important in the study of genes that play a role in the development of cancer and a variety of other single and multiple gene disorders. A wide variety of different methods to detect DNA sequence variations have been developed (Dianzani, 1993). One of these methods, denaturing gradient gel electrophoresis (DGGE), has been shown to be a sensitive, reproducible, reliable technology for use in many research environments (Myers, 1985a; 1985b; 1987).
DGGE involves electrophoresis of double stranded DNA fragments through a polyacrylamide gel containing a linear gradient of denaturant. As the fragment migrates through the denaturant gradient the strands begin to melt. The melting profile of a given DNA duplex is predominantly determined by its base sequence with greater GC content resulting in higher melting temperatures. The ability of DGGE to detect sequence alterations is based on the differential melting characteristics of homoduplex DNA versus heteroduplex DNA. As heteroduplex DNA migrates through the denaturant gradient the areas of non-homology melt at a lower temperature than the comparable homoduplex region. This results in an area of decreased mobility within the fragment, retarding its progress through the gel. This reduction in mobility results in a separation of homoduplex from heteroduplex fragments thereby identifying a region of sequence alteration.
The melting characteristics of a double stranded DNA fragment can be predicted a priori, allowing one to predetermine the conditions under which all alterations within that fragment should theoretically be resolved. The juxtaposition of GC-rich and AT-rich regions commonly seen within the human genome often result in complex melting profiles with adjacent domains demonstrating widely divergent melting temperatures. Such multiple melting domain fragments do not permit the resolution of sequence alterations in the higher melting temperature domains due to the more rapid melting of the lower temperature domains. The addition of a GC rich sequence, (GC clamp), to the 3' or 5' end of an amplicon results in a single melting domain across the fragment, thus allowing the detection of sequence variation across the entirety of the sequence (Myers, 1985c). The prediction of melting domains within a DNA fragment of known sequence has been reduced to a computer algorithm by Lerman and co-workers (Lerman, 1987). The melting profiles determined for the current work were performed using an adaptation of Lermans MELT program created by Bio-Rad, called MacMelt software.
This study presents our efforts within a model system to determine the feasibility of using DGGE in a clinical environment for the detection of mutations in the mismatch repair genes relevant to the development of Hereditary Non-Polyposis Colon Cancer. We selected exon 4 of the hMSH2 gene as a model system for analysis. A series of mutations was created by site directed mutagenesis in an exon 4 containing clone. The individual mutant-containing clones were then analyzed via DGGE to determine the limits of detection of the system.
Materials and Methods
SITE-DIRECTED MUTAGENESIS OF hMSH2 EXON 4
A series of mutations was designed across an amplified region of hMSH2 exon 4. To accomplish the mutagenesis, a series of exon 4 specific PCR primers each with a single base change corresponding to the designed mutation was synthesized. The incorporation of the single base change within the primer resulted in the site-directed mutagenesis of the amplified product. As the mutagenic primers were internal to a larger exon 4 fragment, subsequent amplification of the larger product resulted in an amplified exon 4 containing a known mutation at a known site. The amplified, mutagenized fragment was then cloned into a pGEM-T vector and selected clones were sequenced to verify the presence of the mutation. All subsequent DGGE model experiments were then completed using the cloned material.
The design of DGGE primers follows the general principles of PCR primer design with respect to sequence specificity, lack of internal homology and minimal primer dimer formation. In addition, a GC-rich sequence or clamp is often appended to one of the primers. Such a clamp can be of various lengths and base composition as determined by its ability to produce a single melting domain for the sequence under study. For clinical laboratory use additional criteria include: uniform PCR conditions, amplification of exonic regions inclusive of exon/intron boundaries and a minimum of different DGGE conditions.
The exon 4 model used in this study is an exception to some of these criteria; its purpose was to test the limits of the DGGE assay. To illustrate the general appearance of melt profiles, Figure 1a shows the profile of amplicons with a high divergence from the ideal profile. Figure 1b shows the ideal melting profile achievable with a GC-clamp where the amplicon melts within a single domain.
The polymerase chain reactions consisted of 0.6 micromolar of each primer pair, 250 nanograms of cloned DNA, 10 mM Tris-HCl, pH 8.3, 3 mM MgCl2, 50 mM KCl, 200 M NTPs, 2 units of Taq enzyme, and a 2.5 M final concentration of betaine in a volume of 50 l. The cycling protocol consisted of 32 cycles of 94 degrees for 30 seconds, 55 degrees for 30 seconds, 72 degrees for 1 minute followed by a 5 minute extension at 72 degrees.
Denaturing gradient gels were prepared using a standard gradient mixer and stock gradient mixtures of 6% polyacrylamide-bis-acrylamide (37.5:1) containing denaturant concentrations of 0% urea-formamide in one mixing chamber and 80% urea-formamide in the second mixing chamber. The gradient conditions were calculated from the melting profile as determined by the MacMelt software. The gradient gels were cast and samples were electrophoresed for 5 hours at 130 volts using the DCode universal mutation detection system.
The mutation content of each clone and the DGGE results are described in Table 1. The melting profiles of full length amplicons regardless of the position of the GC-clamp demonstrated extreme fluctuations in melting temperature. The 3' GC-clamped amplicon contained an eight degree variation at its 3' end and a six degree variation at its 5' end resulting in three melting domains (Figure 2a). The complexity of the melting profile using this clamped configuration resulted in no resolvable bands and thus an uninformative analysis. The 5' clamped amplicon contained three melting domains with a ten degree difference at it's 3' end and a 2 degree difference at the 5' end. DGGE detected 7 of the 16 mutations with the majority of the detected mutations located in the lowest melting domain.(Figure 2b) The presence of a HinFI restriction enzyme site 5' to the lowest melting domain allowed the design of two new primers which excluded this lowest melting domain and yielded a more uniform melting profile. The first primer re-design used the original 5' GC-clamped primer and a new non-clamped primer overlapping the HinFI site (termed the 5' HinFI amplicon) (Figure 3a). The second re-design used a non-clamped version of the 5' primer and a 3' GC-clamped primer proximal to the HinFI site (termed the 3' HinFI amplicon) (Figure 3b). A third primer pair was synthesized to include the region of exon 4 excluded by the other primer designs (termed the 3' overlap amplicon) (Figure 3c). DGGE of the 5' HinFI amplicon detected 8 of the 12 mutants known to be present in the amplified product. DGGE of the 3' HinFI amplicon detected 9 of the 12 mutants known to be present. DGGE of the 3' overlap amplicon detected all 5 of the mutations known to be present in the amplicon. Figure 4 depicts a typical gel result following electrophoresis of exon 4 fragments through a DGGE gel. The numbers above the lanes correspond to the mutations detected.
DGGE compares favorably to other gene scanning techniques with respect to detection rate, robustness and size of analyzed fragment. An important advantage of DGGE is the possibility to design primers and gel conditions to maximize the detection of alterations in advance. This is in contrast to other methods of detection that rely on empirical evidence of the discovery of mutations in a fragment. The disadvantages of DGGE include the specialized equipment necessary to run the gels, the consistency of pouring gradient gels in a high throughput environment and the expense of synthesizing clamped primers. In the current model there was no correlation between the position of the mutation and its detection. It is currently accepted, however, that mutations less than 50 bp from the GC-clamped primer are less detectable than those located at a greater distance. The model system demonstrated the importance of melting domains in DGGE design. The 5' GC-clamp amplicon contained three domains with two differing by 10 degrees. The lower temperature domain prevented mutations being detected in the higher temperature domain. The 3' HinFI amplicon also had three domains but two differed by only three degrees. All mutations were detected in the lower two domains. The third domain, which differed by 10 degrees from the lowest melting temperature domain, contained two mutations which were not detected in this amplicon. These results demonstrate the importance of designing amplicons with one or a small number of melting domains which differ by less than 5 degrees. The observed requirement for at least two overlapping amplicons being clamped at opposing ends in these experiments demonstrates an important aspect of DGGE design for the clinical laboratory. To detect the greatest number of mutations, it is often necessary to analyze samples with amplicons clamped in either the 5' or 3' orientations. In this manner it is then possible to obtain scanning data covering an entire amplicon in a single assay. Since this work was completed, a redesigned set of clamped primers is now used to analyze exon 4, reducing the total number of primer pairs from the three used in this work to two sets of GC-clamped primers pairs. Through the use of GC-clamps of various lengths and GC content it is possible to reduce the number of melting domains in virtually all fragments to a single uniform temperature. With careful planning and design of primers and gel conditions, it is possible to analyze large genes in a relatively high throughput manner while maintaining a high degree of sequence alteration detection. Sequencing to determine the precise character of a mutation is then reduced to a single easily analyzed region rather than an entire gene. In summary, DGGE is a robust, reliable technique applicable to both research and clinical endeavors.
1. Dianzani, I., Camaschella, C., Ponzone, A. and Cotton, R. G. H., Dilemmas and progress in mutation detection, Trends in Genetics, 9(12), 403405 (1993).
2. Lerman, L. S., Silverstein, K. Computational simulation of DNA melting and its application to denaturing gradient gel electrophoresis. Methods in Enzymology 155, 482501 (1987).
3. Myers, R. M., Lumelsky, N., Lerman, L. S. and Maniatis, T,. Detection of single base substitutions in total genomic DNA, Science, 313, 495498 (1985a).
4. Myers, R. M., Fischer, S. G., Maniatis, T and Lerman, L. S., Modification of the melting properties of duplex DNA by attachment of a GC-rich DNA sequence as determined by denaturing gradient gel electrophoresis, Nucleic Acids Research, 13(9), 31113129 (1985b).
5. Myers, R. M., Fischer, S. G., Lerman, L. S. and Maniatis, T., Nearly all single base substitutions in DNA fragments joined to a GC-clamp can be detected by denaturing gradient gel electrophoresis, Nucleic Acids Research, 13(9), 31313145 (1985c).
6. Myers, R. M., Maniatis, T. and Lerman, L. S., Detection and localization of single base changes by denaturing gradient gel electrophoresis, Methods in Enzymology, 155, 501527 (1987).
back to top