In 1977 Frederick Sanger described a method of DNA sequencing using chain-terminating nucleotides called dideoxynucleotides.
The aim was to determine the sequence of nucleotides in a piece of DNA using these artificial or synthetic nucleotide analogues. Unlike natural DNA nucleotides (deoxyribonucleoside triphosphates, dNTPs), these lab-synthesised forms lack the 3'-hydroxyl group required for DNA strand extension.
This method became known as Sanger sequencing.
These chain-terminating nucleotides are called dideoxyribonucleoside triphosphates (ddNTPs).
DNA is made up of a chain of four different nucleotides called dNTPs. To copy DNA and extend a DNA strand, DNA polymerase adds a complementary nucleotide.
A closer look at its structure shows that a dNTP consists of a deoxyribose sugar, a nitrogenous base, and a triphosphate group. A nucleoside consists of a sugar and a base. In DNA the sugar is deoxyribose, while in RNA it is ribose. The base is one of four bases: adenine, thymine, guanine, or cytosine.
A ddNTP lacks both the 2'-OH and 3'-OH groups found in ribose. Compared with deoxyribose, it is missing the 3'-OH group.
The role of DNA polymerase is to add new nucleotides to a growing DNA strand. During DNA synthesis, the 3'-hydroxyl group (3'-OH) of the growing DNA strand reacts with the α-phosphate of the incoming dNTP, forming a phosphodiester bond and releasing pyrophosphate.
If a ddNTP is incorporated into the DNA strand, synthesis stops because the ddNTP lacks the 3'-OH group required to add the next nucleotide. This absence of a 3'-OH group terminates DNA chain elongation.
It is also useful to understand the naming convention of 5' (five-prime) and 3' (three-prime). The carbons in the deoxyribose sugar are numbered 1' through 5'. The nitrogenous base is attached to the 1' carbon, while the phosphate group is attached to the 5' carbon.
The 3'-OH group attached to the 3' carbon is the chemical group required for DNA strand extension. Because DNA polymerase adds new nucleotides to the existing chain utilising the phosphate group of the new dNTP. Hence, the dogma that DNA synthesis proceeds in the 5'→3' direction. And when DNA sequences are written, they are conventionally written from 5' to 3'.
DNA polymerase adds nucleotides complementary to the template strand, so C pairs with G and A pairs with T.
So how does Sanger sequencing work?
The original Sanger sequencing method was different from the one used today. The original method was completely manual and used radioactive labels.
Let's take a look at the original Sanger sequencing method.
We need a primer, DNA polymerase, dNTPs, a DNA template, and ddNTPs.
One of the dNTPs, usually dATP, is labeled with a radioactive isotope.
A total of four tubes are used, one for each ddNTP.
To begin, the DNA template is heated to denature the double-stranded DNA into single strands. Remember, this was before PCR existed. Because the DNA polymerases available at the time were not thermostable, the enzyme was added after the denaturation step.
The mixture is then cooled to allow the sequencing primer to anneal to the template.
DNA polymerase, all four dNTPs, and one of the four ddNTPs are then added to each tube.
DNA polymerase extends the primer along the DNA template. Occasionally, a ddNTP is incorporated instead of a dNTP, terminating the DNA fragment.
Because the ddNTP is present at a much lower concentration than the corresponding dNTP, incorporation occurs randomly.
The result is a collection of DNA fragments that terminate at every occurrence of that particular base, generating fragments of different lengths.
All fragments in a tube begin with the same primer sequence and end with the same terminating nucleotide.
Low incorporation of the ddNTP allows longer stretches of DNA to be sequenced.
In the original Sanger method, read lengths of approximately 200 nucleotides were achievable.
Next, the sequencing reactions are mixed with loading dye and loaded into separate lanes of a polyacrylamide gel.
The fragments migrate through the gel according to size, with smaller fragments moving faster than larger fragments.
Polyacrylamide gels have sufficient resolution to distinguish DNA fragments that differ by a single nucleotide in length.
At this stage the fragments cannot be seen.
The loading dye indicates when the fragments have migrated through the gel.
To visualise the DNA fragments, the gel is dried onto a support and exposed to X-ray film.
The radioactive labels incorporated into the DNA fragments expose the film, producing a pattern of bands.
The process of determining the DNA sequence from these bands is called base calling.
The gel is read from the bottom upward, starting with the shortest fragment. This reveals the sequence of the newly synthesized DNA strand in the 5'→3' direction.
For example, if the shortest fragment appears in the ddTTP lane, the first base called is T. If the next shortest fragment appears in the ddGTP lane, the next base is G.
Continuing upward through the gel reveals the complete sequence.
The original Sanger sequencing method was very labor-intensive. It could take several days to generate approximately 200 nucleotides of sequence from only a small number of samples.
There was a strong need to streamline and automate the process.
Applied Biosystems created the first commercial automated DNA sequencing instrument in 1987, the AB370A.
Researchers had already demonstrated that fluorescent dyes could replace radioactive labels. These fluorescent dyes were safer and eliminated the need for time-consuming X-ray film detection.
In this instrument, each of the four sequencing reactions was labeled with a different fluorescent dye.
After the sequencing reactions were completed, all four reactions could be combined and loaded into a single lane of a gel.
The AB370A used a laser to detect fluorescent DNA fragments as they migrated through the gel.
The instrument automatically transferred the data to a computer, which performed automated base calling.
Up to 16 samples could be run simultaneously, with read lengths approaching 450 nucleotides.
The AB370A demonstrated that DNA sequencing could be faster and more automated.
Scientists began to think that sequencing the entire human genome might be achievable.
In 1990 the U.S. government launched the Human Genome Project, an international effort to map and sequence the entire human genome.
Sequencing the human genome promised major advances in biology and medicine, including identifying genes associated with inherited diseases and improving our understanding of human biology.
Kary Mullis had invented PCR in 1983, but it was not until 1989 that Vincent Murray applied thermostable Taq polymerase to Sanger sequencing.
In traditional Sanger sequencing, most labeled primers remain unused because the primer is present in excess relative to the DNA template.
The use of Taq polymerase allowed repeated cycles of denaturation, primer annealing, and extension, similar to PCR.
Because only a single sequencing primer is present, newly synthesised strands do not serve as templates for exponential amplification.
As a result, the amount of sequencing product increases approximately linearly rather than exponentially. This process became known as cycle sequencing.
The increased signal generated by cycle sequencing also reduced the amount of input DNA required.
Another important advance was capillary electrophoresis.
In capillary electrophoresis, DNA fragments migrate through a thin capillary filled with a polymer matrix under an electric field.
The narrow capillary efficiently dissipates heat, allowing higher voltages to be used without overheating.
Higher voltages result in faster separations and improved resolution.
Beckman Coulter launched the first commercial capillary electrophoresis instrument in 1989.
This technology paved the way for capillary-based Sanger sequencing systems.
Applied Biosystems launched the ABI Prism 310 in 1995, marking the beginning of modern Sanger sequencing.
The ABI Prism 310 replaced slab gels with a single capillary.
A sequencing run could be completed in under three hours, with read lengths approaching 600 base pairs.
The system also automated sample loading and reduced DNA input requirements through electrokinetic injection.
DNA fragments were separated by size, detected by a laser, and analyzed automatically by software that performed base calling.
Although fluorescently labeled ddNTPs were available, fluorescent primer labeling was initially preferred because it produced more uniform signal intensities.
This changed with the introduction of BigDye Terminator chemistry in 1997.
BigDye Terminators improved the balance of fluorescent signal intensity among dye-labeled ddNTPs, allowing all four termination reactions to be performed in a single tube.
This greatly simplified sequencing workflows.
The Human Genome Project continued to drive demand for faster and more automated sequencing technologies.
In 1998 Applied Biosystems launched the ABI Prism 3700, which contained 96 capillaries.
The ABI Prism 3700 played a major role in sequencing the human genome.
Each run processed 96 samples simultaneously, generated read lengths approaching 800 base pairs, and required minimal hands-on time.
Celera Genomics, led by Craig Venter, purchased hundreds of ABI Prism 3700 instruments and used them to compete directly with the publicly funded Human Genome Project.
Celera produced a draft human genome sequence in 2001, and the Human Genome Project published its own draft sequence the same year.
As an aside, the Human Genome Project did not sequence the DNA of a single individual. The reference genome was assembled from DNA obtained from multiple anonymous donors. Much of this DNA came from blood samples (while red blood cells lack nuclei and therefore contain no genomic DNA, white blood cells contain nuclei and provide a rich source of genomic DNA). The resulting reference genome was therefore a composite sequence rather than the genome of any one person.
Modern Sanger sequencing remains widely used today.
Sanger sequencing typically achieves greater than 99.9% accuracy in high-quality reads, which is why it is often considered the benchmark against which other sequencing technologies are compared.
For small projects involving individual genes, plasmids, or a limited number of samples, Sanger sequencing is often faster and more cost-effective than next-generation sequencing (NGS).
However, Sanger sequencing has lower throughput and lower sensitivity for detecting rare variants, typically requiring variants to be present at roughly 15–20% frequency before they can be reliably detected.
In contrast, NGS can detect variants present at much lower frequencies and can generate billions of reads and terabases of sequence data in a single run.
This allows many whole human genomes to be sequenced simultaneously.
For validating variants, sequencing plasmids, or analysing a small number of genes or samples, Sanger sequencing remains one of the most practical and widely used sequencing methods available.
Related post:
What Is Next-Generation Sequencing? NGS vs Sanger Explained
https://adwoabiotech.blogspot.com/2026/06/what-is-next-generation-sequencing-ngs.html
Oxford Nanopore Sequencing: A Different Philosophy of Reading DNA
https://adwoabiotech.blogspot.com/2026/06/what-is-oxford-nanopore-sequencing-how.html
