Danish Center for Human Genome Research, The John F.Kennedy Institute, GI. Landevej 7,. DK-2600 Glostrup, Denmark. Received March 22, 1993; Accepted ...
1.1 1993 Oxford University Press
Nucleic Acids Research, 1993, Vol. 21, No. 9 2261 -2262
A simple method for identification of point mutations using denaturing gradient gel electrophoresis Per Guldberg and Flemming Guttler Danish Center for Human Genome Research, The John F.Kennedy Institute, GI. Landevej 7, DK-2600 Glostrup, Denmark Received March 22, 1993; Accepted April 1, 1993 Denaturing gradient gel electrophoresis (DGGE) (1) is a fast and reliable method for detection of single base alterations in fragments of DNA. In combination with PCR, DGGE has become one of the most widely applied methods for detection of point mutations in human genes. In the most common approach, mutations are located to limited regions of the gene by DGGE and subsequently identified by sequence analysis. It is, however, often desirable to circumvent sequencing by deciding whether a gene alteration is one of previously established mutations. A variety of PCR-based diagnostic methods for detection of known mutations have been developed, but for each mutation these methods all require a set of oligonucleotides for hybridization or amplification. The mutation resolving capacity of DGGE is based on the fact that similar DNA fragments differing only by a minute sequence alteration, such as a single base substitution, migrate differently in a denaturing gradient gel due to differences in melting characteristics. Increased electrophoretic resolution can be achieved by analysis of heteroduplexes (2) which are formed during PCR when the target DNA contains two different alleles. Heteroduplex molecules carry a single base mismatch and will, due to their lower thermostability, be retarded in the gel at a lower concentration of denaturants than the corresponding homoduplexes. Since each base substitution has a particular influence on the melting properties of a DNA fragment (3), each mutation is characterized by a 'fingerprint' on the gel in the form of a composite pattern of homoduplex and heteroduplex bands, providing indication of the identity of the mutation. Two different mutations can, however, result in similar band images, and further identification is therefore necessary. Here we show that such identification can be accomplished simply by mixing the test sample with a control sample containing a known mutation, followed by heteroduplex formation and DGGE. If the two mutant fragments are not identical, additional mismatched heteroduplexes will be formed by random reassortment of the different singlestrand constituents and revealed as novel bands on the gel. The practical approach is demonstrated for a 23 1-bp GCclamped (4) fragment encompassing exon 12 of the human phenylalanine hydroxylase gene (5). The melting map of this fragment and the positions of six previously identified mutations are shown in Figure 1. All mutations could readily be resolved as heteroduplexes by DGGE, but two mutations (Y414Y and D415N) resulted in indistinguishable band patterns (Figure 2A). To establish whether the mutation in a test sample was one of the known mutations, the test sample was mixed with each of the control samples. Heteroduplexes were generated by heating
9590-
R408W
R4080
P
85-
BS.
80-
g
75-
Y414C E Y41 4Y lD41 5N \
~~~~~~~~~IVS-12ntl
70-
65-
EXON 12
vv-r
100 150 Base number
50
0
200
250
Figure 1. Melting map of a GC-clamped fiagment covering exon 12 of the human PAH. Arrows indicate positions of previously established mutations.
C) c)e E m ::
.,2
f GO z3
> cr-
H
--
HHHHHH t
H
A
B
Figure 2. DGGE of the PAH fragment. 15 IeI of amplified product was electrophoresed through a 20-65% denaturant, 6% acrylamide gel for 7 h at 60°C and 150 V. (A) Band images of a test sample harbouring an unknown mutation, wild-type sample and six samples from individuals heterozygous for known mutations. (B) 7.5 A1 of the test sample was mixed with 7.5 u1 of each of the seven control samples. Heteroduplexes were generated by boiling for 5 min, and subsequent reannealing in two 1 h steps at 65°C and 37°C, respectively. Heteroduplexes were analysed on a new DGG under the same conditions as described above. WT = wild-type.
2262 Nucleic Acids Research, 1993, Vol. 21, No. 9 and subsequent cooling, and the samples were analyzed on a new denatuing gradient gel (Figure 2B). In five of the mixed samples, additional heteroduplex bands appeared as compared to the test sample, stating the presence of more than two alleles. In the sample containing a fragment harbouring the D415N mutation, the band pattern remained unchanged, identifying the mutation in the test sample as D415N. The band pattern also remained unchanged in the sample containing two wild-type alleles, confirming that the second allele in the test sample was wildtype. In standard procedures, the control samples to be mixed with the test sample can be limited to samples displaying a band pattern similar to the test sample. The present approach does not require primers or materials additional to those used for the mutation detection procedure, providing a simple, rapid and inexpensive method for mutation identification.
ACKNOWLEDGEMENTS The technical assistance of Karen F.Henriksen is acknowledged. P.G. is supported by Fellowship 99-9903 from the Danish Research Academy. This study is supported in part by The Danish Medical Research Council, The Danish Biotechnological Research and Development Programme 1991-1995, The Danish Health Insurance Foundation, and The Novo Foundation.
REFERENCES 1. Myers,R.M., Maniatis,T. and Lerman,L.S. (1987) Methods Enzymol. 155, 501-527. 2. Myers,R.M., Lumelsky,N., Lerman,L.S. and Maniatis,T. (1985) Nature 313, 495-497. 3. Imman,L.S., Fischer,S.G., Hurley,I., Silverstein,K. and Lumelsky,N. (1984) Ann. Rev. Biophys. Bioeng. 13, 399-423. 4. Sheffield,V.C., Cox,D.R., Lerman,L.S. and Myers,R.M. (1989) Proc. Natl. Acad. Sci. USA 96, 232-236. 5. Kwok,S.C.M., Ledley,F.D., DiLeIJa,A.G., Robson,K.J.H. and Woo,S.L.C. (1985) Biochemistry 24, 556-561.
