Megabase Sequencing

Megabase Sequencing

In addition to mapping, sequencing often provides fundamental infor-mation for further studies on a gene or gene system. The sequencing techniques described are adequate for sequenc-ing a few genes, but one area of great interest, the immune system, possesses hundreds of genes, many of unknown function. Here hun-dreds of thousands of nucleotides must be sequenced. Serious effort is now also going into the early steps of determining the sequence of the entire human genome. Large sequencing projects such as these require better methods, and several have been developed. The one described below eliminates the use of radioisotopes and automates the detection of the bands on gels.

A number of steps in the standard DNA sequencing procedure seri-ously limit data acquisition. These are obtaining the plasmids necessary for sequencing the desired region, pouring the gels, exposing and developing the autoradiograph films, and reading the information from the films. Several of these steps can be streamlined or eliminated.

Imagine the savings in the Sanger sequencing technique if each of the four dideoxynucleotides could be tagged with a unique label. Then, instead of labeling the primer or the first nucleotides synthesized, the chain terminating nucleotide would possess the label. If this were done and each of the four labels were distinguishable, the four dideoxynu-cleotides could be combined in the same synthesis tube and the complex mixture of the four families of oligonucleotides could be subjected to electrophoresis in the same lane of the gel. Following electrophoresis, the four families of oligonucleotides could be distinguished and the entire sequence read just as though each one occupied a unique lane on the gel.

Figure 10.13 Excitation and emission spectra suitable for DNA sequencing.Each of the four fluorescent groups that emits at wavelengths λ1, λ2, λ3, and λ4, would be attached to a different base.

Instead of using radioactive label, a fluorescent label is used. In order that this approach work well, the fluorescent adduct on the dideoxynu-cleotide must not interfere with the nucleotide’s incorporation into DNA. Furthermore, each of the four nucleotides must be modified with a different adduct, one that fluoresces at a different wavelength from the others. In addition, it is useful if the excitation spectrum of the four fluorescent molecules substantially overlap so that only one exciting wavelength is required (Fig. 10.13).

Although the entire gel could be illuminated following electrophore-sis, it is easier to monitor the passage during electrophoresis of one band after another past a point near the bottom of the gel. By measuring the color of the fluorescence passing a point near the bottom of the gel, the nucleotide terminating this particular size of oligonucleotide can be determined (Fig. 10.14). One after another, from one nucleotide to the next, the oligonucleotides pass the illumination point and the color of their fluorescence is determined, yielding the sequence of the DNA. Multiple lanes can be monitored simultaneously so that the sequence can be determined semiautomatically of many different samples simul-taneously. Each lane of such a gel can provide the sequence of about 400 nucleotides of DNA.

The sensitivity of such a DNA sequencing approach is less than the radioactive techniques, but it is still sufficiently high that small DNA samples can be successfully used. A more serious problem than the sensitivity is the generation of useful samples to be sequenced. One approach is to generate many random clones from the desired DNA in a vector suitable for Sanger sequencing, to sequence at least the 300 nucleotides nearest to the vector DNA, and then to assemble the se-quence of the region by virtue of the overlaps between various se-quences. This shotgun approach yields the desired sequences if sufficient clones are available and sufficient time and effort are ex-pended. In the sequencing of any sizeable amount of DNA, a pure shotgun approach is not efficient, and great effort is required to close the “statistical” gaps. When a few gaps remain, it may be easier to close them by chromosome walking than by sequencing more and more randomly chosen clones, most of which will be of regions already sequenced.

Figure 10.15 Generation of deletions for sequencing by exonuclease digestion.The vector is opened and digested, then a fragment is removed by cutting a second time with a restriction enzyme. This fragment is recloned and sequence is determined by using a primer to a sequence within the cloning vector adjacent to the location of the inserted fragment. By performing a series of exonuclease digestions for increasing periods, progressively larger deletions may be ob-tained.

Another method of generating the necessary clones for sequencing a large region is to use a nested set of overlapping deletions. By sequenc-ing from a site within the vector sequences with the use of an oligonu - cleotide that hybridizes to the vector, the first 400 or so nucleotides of each of the clones can be determined. The resulting sequences can easily be assembled to yield the sequence of the entire region.

Such a set of clones can be generated by opening a plasmid containing the cloned DNA, digesting with an exonuclease for various lengths of time, and recloning so that increasing amounts of the foreign DNA inserted in the plasmid are deleted (Fig. 10.15).