The process of determining the order of nucleotides (A, C, G, and T) along a DNA strand is called DNA sequencing. Knowing the nucleotide sequence of a gene or region of DNA is important in studying relatedness between species and between individuals and for a better understanding of how genes function. Several techniques have been developed for "reading" the sequence of any particular DNA segment. One of these techniques was developed by Fred Sanger in 1977 and is called the chain termination method (or Sanger method). The essence of the technique is the creation of a set of DNA fragments that match the chain to be sequenced. Each fragment is one nucleotide longer than the last. By determining the identity of the final nucleotide in each fragment, the sequence of the whole chain can be determined.
The chain termination method makes use of special forms of the four nucleotides that, when incorporated at the end of a growing chain during DNA synthesis, stop (terminate) further chain growth. In four separate reactions, each containing a different terminator base (called a dideoxynucleotide), a collection of single-stranded fragments is made. These fragments all differ in length and all end in the dideoxynucleotide added to the particular reaction. Gel electrophoresis is then used to separate the fragments according to their length. By knowing which terminator base is associated with which fragment on the gel, the base sequence can be constructed.
The Need for Automated Sequencing
When chain termination sequencing is performed manually, each of four reaction tubes contains a different type of terminator base as well as a radioactive nucleotide for labeling the DNA fragments as they are made. Each of the four reactions is electrophoresed in a separate lane of a gel, and X-ray film is used to detect the fragments. Using this technique, a dedicated and skilled technician can determine the sequence of as many as 5,000 bases in a week. Demand for the ability to read more sequence in a shorter amount of time, however, led to the development of instruments that could, with the aid of computers, automate the DNA sequencing process.
The first step toward this goal was achieved in 1985, when Leroy Hood at the California Institute of Technology attached fluorescent dyes to the primer used in the sequencing reactions; each different color dye (blue, green, yellow, and red) was matched with a different terminator base. He and Michael Hunkapiller from Applied Biosystems, Inc. (ABI) built an instrument, dubbed the ABI Model 370, to read the sequence of the dyelabeled fragments. It was equipped with an argon ion laser for exciting the dyes, a flat gel laid between two glass plates (referred to as a "slab" gel) capable of sixteen-lane electrophoresis, and a Hewlett-Packard Vectra computer boasting 640 megabytes of memory for data analysis.
Using fluorescent dyes, all four sequencing reactions could now be loaded into a single gel lane. As the fragments electrophoresed, the beam of the laser focused at the bottom of the gel made the dye-labeled fragments glow as they passed. The color of each dye-labeled fragment was then interpreted by the computer as a specific base (A if green, C if blue, G if yellow, and T if red). Over 350 bases could be read per lane. With this new automated approach, a technician could read more sequence in a day than could be read manually in an entire week.
Refinements in Automation
Shortly after ABI placed its automated DNA sequencer on the market, the Dupont company introduced its own model, the Genesis 2000. Dupont had also developed a new method of labeling sequencing fragments: attaching the fluorescent dyes to the terminator bases. With this innovation, four separate sequencing reactions were no longer required; the entire sequencing reaction could be accomplished in a single tube. However, Dupont failed to effectively compete in the automated sequencer market and sold the rights to the dye terminator chemistry to ABI.
ABI continued to refine its automated sequencer. More powerful computers, increased gel capacity (to 96 lanes), improvement of the optical systems, enhancement of the chemistry, and the introduction of more sensitive fluorescent dyes increased the reading capacity of the instrument to over 550 bases per lane. The ABI PRISM Model 377 Automated Sequencer, introduced in 1995, incorporated these changes and could read, at maximum capacity, over 19,000 bases in a day. Even at this rate, however, the sequencing of entire genomes, as that of humans (3 billion bases in length), was still not practical. Genome sequencing awaited several further innovations.
Working with the Model 377 Automated Sequencer, a laboratory technician had to pour the slab gels and mount them on the instrument. This process alone was time-consuming and cumbersome. In addition, the technician had to add each sequencing reaction into each individual lane of the gel prior to the run. The MegaBase, developed by Molecular Dynamics, and the ABI Model 3700 Automated Sequencer, developed by ABI, addressed these limitations by using multiple capillaries, thin, hollow glass tubes filled with a gel polymer.
The ABI PRISM Model 3700 Automated Sequencer, developed with the Hitachi Corporation and having a price tag of $300,000, uses ninety-six capillaries, each not much wider than a strand of human hair. The capillaries are automatically cleaned and filled with fresh gel polymer between each electrophoresis run. The instrument is also equipped with a robot arm that automatically loads the sequencing reactions into the capillaries, greatly decreasing the amount of human labor required for its operation. The Model 3700 Automated Sequencer can read over 400,000 bases in a day, a greater than twenty-fold increase over the maximum capacity of the Model 377. Beginning in September 1999 and using 300 of these instruments, the Celera Corporation had sequenced the entire human genome five times over within four months.
see also Cycle Sequencing; Gel Electrophoresis; Nucleotide; Sanger, Fred; Sequencing DNA.
Frank H. Stephenson
and Maria Cristina Abilock
Smith, Lloyd M., et al. "Fluorescence Detection in Automated DNA Sequence Analysis." Nature 321 (1986): 674-679.