Australasian Biotechnology (backfiles) AusBiotech ISSN: 1036-7128 Vol. 6, Num. 5, 1996

Allele-specific PCR

Edward Edkins PhD

Senior Scientist Clinical Chemistry Joint Women's and Children's Hospital, Box D184, GPO Perth 6001, Western Australia. Phone (09)340:8595. Fax (09)340:8117 Email: tedkins@uniwa.uwa.edu.au

Code Number: AU96011
Size of Files:
    Text: 8.2K
    Graphics: No associated graphics files

Summary

Performance in a clinical setting of cystic fibrosis screening using allele-specific PCR. The advantages of multiplex PCR and the difficulties of routine screening for up to fourteen different mutant alleles are summarised.

The basic polymerase chain reaction has been modified in a number of ways. In the basic method two primers are picked such that the area of DNA of interest is amplified and the resulting product used in a number of ways. In a diagnostic molecular genetics laboratory a common aim is to look for mutations in the PCR product and this can be accomplished by looking for a loss, or gain, of a restriction enzyme site, making a dot blot of the product and probing with a labeled oligonucleotide or measuring, by polyacrylamide electrophoresis, the size of the product (Haliassos et al. 1989, Rommens et al. 1990). Each of these methods have advantages and disadvantages in a routine laboratory but the main disadvantage is in the time and labour needed to obtain a result. Some years ago a quicker method called allele-specific PCR was invented which had advantages as to speed and cost over the more conventional methods (Newton et al. 1989).

In this method one of the primers of the reaction was moved up to the mutation site such that the 3' end of the primer sat right at the mutation. It was noted that primers were very sensitive as to the correct sequence at the 3' end and would not tolerate mismatches if a product was to be made. Primers would, however, tolerate quite large sequence variation at sequences away from the immediate 3' end. A method was devised such that two PCR reactions were performed in which a common primer upstream or downstream of the mutation was used together with primer in one tube which matched the wild-type sequence and in the other a primer matching the mutation was used. If only the wild-type sequence existed in the DNA sample being amplified then a product was obtained in the normal tube and no product was obtained in the mutant tube. Conversely, if only mutant sequences were in the DNA sample then a product was obtained in the mutant tube and none in the normal tube. It is clear that if the DNA was heterozygous at the locus in question then a product was obtained in both tubes.

Using this method only 2 PCR reactions per DNA sample was required followed by simple electrophoresis in an agarose gel and staining with ethidium bromide. From a viewing of the bands the presence or absence of the mutation could quickly be seen.

As the method has been explained it will be seen to have a major flaw. The absence of a band in one or the other of the tubes will be interpreted as an absence of that particular sequence, whether wild type or mutant. There is the possibility that the absence of a band is due to the failure of that PCR reaction and not the absence of the sequence. To overcome this problem a control set of primers are included in each of the two tubes picked in a such a way that a band is always seen with this set of primers. Often the control primers amplify a sequence in another part of the genome and are picked to be able to be amplified under the same conditions of PCR and in having a size different to the bands of the mutational set.

Multiplexing

In our laboratory we have use the allele-specific PCR to detect the presence of 8 of the common mutations causing cystic fibrosis (CF). The gene in question is the CFTR gene and one mutation, the loss of 3 bases in exon 10, accounts for 75% of the mutations in CF chromosomes. A handful of other mutations bring this figure up to around 85% while over 600 mutations have been found by various laboratories around the world. Our laboratory will screen for some 14 mutations but in routine work limit this search to the 8 most common.

To make this method even more cost effective the primers can be multiplexed, that is to search for more than one mutation per set of two tubes. We have followed the method of Ferrie et al (1992) in using four primers sets per tube so to complete the investigation of the 8 mutations only four PCR reactions are needed. This makes for a very quick and cheap method and allows the turnaround of results from the laboratory to be carried out expeditiously.

One problem arises with adding more primer sets to each tube in that one of the primer pairs can take over the reaction and no products can be obtained from the other sets. This can usually be overcome by adjusting the concentrations of primers such that a balanced set of bands result. Sometimes a ten-fold difference in concentrations are required to balance up the reactions. An advantage in multiplexing the primers is that the control set of primers can be dispensed with. With four sets in each reaction, two normal and two mutant in each tube at least one band will be seen in each tube. This tells the scientist that the reaction must have worked in that tube and so acts as the control. With less than four sets or with all the normals in one tube and all the mutants in the other tube this is not accomplished and a control set of primers will again be needed.

As the results are to be read from an agarose gel each band, corresponding to a mutant or wild type sequence, will have to be of a different size. This is easily accomplished by setting the common primer at a suitable position to the testing primer. It is also sensible to make the products of a suitable size for easy amplification and this means picking sizes in the 100 to 600 base pair range. Smaller sizes are not easily seen on agarose gels and larger sizes can be difficult to amplify without modification to the PCR reaction.

The method relies, as described before, in a mismatch at the 3' end of one of the primers. The question can be asked how much mismatch is required in the primer? In general only the most 3' position needs to be mismatched for a product not to be formed but in practice some futher mismatches may have to be introduced. Ferrie et al (1992) give a table in their paper in which further mismatches are introduced at the penultimate base of the primer depending on the base at the end of the primer. Using these mismatches primers can be developed in which no product can be obtained under normal PCR conditions if the particular sequence is not present in the sample.

The method set out in this paper has been used in our laboratory since 1993 and has proved itself to be robust and reliable in routine use. If the quality of DNA is not good some problems do occur such as no bands appearing or weak bands. Improving the quality of sample nearly alway overcomes these rare problems.

References

Ferrie RM, Schwarz MJ, Robertson NH, Vaudin S, Super M, Malone G, Little S (1992) Development, multiplexing, and application of ARMS tests for common mutations in the CFTR gene Am. J. Hum. Genet 51 251

Haliassos A, Chomel JC, Tesson L, Baudis M, Kruh J, Kaplan JC, Kitzis A (1989) Modifications of enzymatically amplified DNA for the detection of point mutations. Nucl Acids Res 17 3606

Newton CR, Graham A, Hepinstall LE, Powell SJ, Summers C, Kalsheker N, Smith JC et al (1989) Analysis of any point mutation in DNA: the amplification refractory mutation system. Nucl Acids Res 17 2503

Rommens JM, Iannuzzi MC, Kerem B-s, Drumm ML, Melmer G, Dean M, Rozmahel R et al (1989) Identification of the cystic fibrosis gene. Science 245 1059

Copyright 1996 Australian Biotechnology Association Ltd.