DNA Sequencing: the Sanger Method
Dr. Steven Vik
4 December 2001
DNA Sequencing: the Sanger Method
Frederick Sanger was born on August 13, 1918. His father was a medical practitioner, and when considering what field to enter, Sanger initially intended to study medicine. Once he began university at Cambridge, Sanger developed an interest in biochemistry. He earned his B.A. from Cambridge and remained there to focus on his Ph.D. Eventually Sanger became a professor at Cambridge, doing his research in proteins, including insulin. In 1958 he received a Nobel Prize in Chemistry for the sequencing of insulin, which was the first protein to be completely sequenced. Sangers interest then shifted from the sequencing of amino acids to the sequencing of nucleic acids.
In 1977, Sanger described a new DNA sequencing method, which is now more widely used than the older Maxam and Gilbert method. He, along with Paul Berg and Walter Gilbert, received a Nobel Prize in 1980 for the new method, which relies on DNA polymerases "ability to use 2',3'-dideoxynucleoside triphosphates [ddNTPs] as substrates" (3, p.170). At the 3' end of ddNTPs, there is a hydrogen instead of the hydroxyl group necessary for DNA chain elongation. This altered structure causes termination of the growing DNA chain and is the basis behind Sangers method, which is also called the dideoxy method.
The DNA sequencing reaction begins with a single stranded DNA molecule for which an oligonucleotide primer has been designed. For the sequencing of PCR products, which are double stranded, the DNA is heat denatured and then rapidly frozen by placing it in liquid nitrogen or dry ice. This prevents the separate strands from re-annealing. The primer anneals to the corresponding sequence on the single stranded DNA and DNA polymerase directs the synthesis of a complementary copy of the template from the 3' end of the primer. Several different DNA polymerases are appropriate to use for this technique, including T7 DNA polymerase, Bst DNA polymerase I, Pfu exo-, and Taq. The reaction is carried out in four different tubes, one for each nucleotide. In each tube the template DNA, DNA polymerase, and all four nucleotides are added. The end products must be labelled in order to be detected, and
this is done by attaching radioactive labels to one of the dNTPs used in the reaction. Several radioactive labels can be used, including ["-32P], ["-33P], and ["-35S]. Usually, ["-33P] is used because it gives better resolution than ["-35S] and is safer to use than ["-32P]. A relatively small amount of one type of ddNTP is also added to the appropriate tube (eg. ddATP to tube 1, ddGTP to tube 2, etc.).
During the reaction in, for example, tube 1, each time the DNA polymerase reaches a T position on the template, there is a small chance that ddATP will be incorporated, terminating the chain elongation. In most cases, the DNA polymerase will add dATP and the DNA synthesis will continue until another T position is reached, where there will be another small chance for chain termination to take place. This occurs in each tube until the end of the reaction. Every possible fragment length will be synthesized, with the shortest chain being the length of the primer plus one nucleotide.
The reaction products are separated, based on size, using a thin polyacrylamide gel, which usually contains 7M urea. The urea "acts as a denaturant to reduce the effects of DNA secondary structure" (2, p.32). The high temperature that the gel is run at also contributes to the denaturation of the DNA. Keeping the DNA denatured is very significant as the fragments being separated differ in length by just one nucleotide. The use of polyacrylamide gels is also important, because their small pore size can resolve DNA to just one base. Four lanes are used, one for each tube used in the reaction. Once the gel has been run, it is exposed to X-ray film, resulting in an autoradiogram which can be read from the bottom up, giving the 5' to 3' sequence of the DNA that was generated in the reaction. Up to 400 or 500 bases can be sequenced using this manual method.
There is an automated version of DNA sequencing which involves the use of fluorescent tags instead of radioactive labels. These tags can be attached to primers that are made with a 5'-amino group. Four different fluorescent dyes are used, one for each of the four reactions (which still must be done in separate tubes). This way each nucleotide reaction can be identified by a
unique fluorescence wavelength. This method gives the most uniform sequence data, but the dyes can also be incorporated into the ddNTPs to allow any primer to be used for the sequencing reaction. When using fluorescent ddNTPs, a different dye is attached to each different ddNTP giving each a unique wavelength emission. When tagging the ddNTPs, all four reactions may be carried out in one tube.
The separation of fluorescent tagged sequencing products is based on size, and can be done using either a polyacrylamide gel or a capillary electrophoresis system. In the case of the gel, the reaction products may all be run in one lane and are automatically read as they electrophorese past a detector that reads the wavelength emitted by each band. The capillary system consists of a thin-coated capillary that contains nonpolymerized gel matrices through which the fragments move, where they are again automatically detected by a detector. After the bands pass the detector, they continue into a buffer chamber. These systems allow longer sequences to be read because the gel or capillary is allowed to run longer. The capillary system is almost completely automatic, as the application and removal or the polymer from the capillary, both the loading and the running of the samples, and the detection of the products are all automated.
The manual and automated sequencing reactions described above both come from Sangers method of using dideoxy nucleotides. They both rely on the occasional termination of chain elongation due to the lack of the 3' hydroxy group, and they both require detection methods and separate the products based on size. The automated method reduces the amount of work involved in the process, enabling the use of just one reaction tube (when the ddNTPs are labelled) and by automating most of the sequencing procedure. It also lowers the chance of human error by relying on laser detection and not the time-consuming process of reading an autoradiograph by hand. The newer, automated techniques are expensive, but can also read many more sequences in a shorter time period because up to 100 columns may be run and read by the detector simultaneously. The sequences read using the lasers will also be longer than
those read by hand, which are limited to the length of the gel. In conclusion, there is a very fast, efficient, yet costly way of sequencing. There is also a more time consuming and demanding technique that is more practical for many laboratories that do not own the expensive automated equipment. Overall, though, both methods depend on Sangers original principles and will give the researcher their desired answer, which is the sequence of their DNA.
1. McPherson and Møller, PCR. BIOS Scientific Publishers: Oxford; 2000
2. Nicholl, An Introduction to Genetic Engineering. Cambridge University Press: Cambridge; 1994.
3. Old and Primrose, Principles of Gene Manipulation, 5th ed. Blackwell Science: Oxford; 1994.
4. Sanger, Frederick. "Autobiography of Frederick Sanger." 1980. http://www.nobel.se/
5. "Sanger, Frederick." Microsoft Encarta Online Encyclopedia 2000. index/conciseindex/5B/05B8A000.htm?z=1&pg=2&br=1"http://encarta.msn.com/
6. University College of London, Molecular Biology course (BIOLB236) handouts. London;