Since the 1950s, scientists have discovered a broad variety of new techniques for medical analysis and testing. Where scientists once searched for disease-related antibodies in a patient, they can now identify the disease organism itself by methods not previously imagined. And they can focus on the organism itself rather than on evidence that it once was there.

Central to these techniques is the DNA molecule. It is now possible to reproduce DNA in a test tube, fragment it, determine its composition, change its structure, and map its genes. The principles learned from these breakthroughs have been applied in diagnosing infectious disease, screening individuals for cancer and genetic diseases, and ensuring the public health by identifying pathogens in the environment.

Central to these techniques is the DNA molecule. It is now possible to reproduce DNA in a test tube, fragment it, determine its composition, change its structure, and map its genes. The principles learned from these breakthroughs have been applied in diagnosing infectious disease, screening individuals for cancer and genetic diseases, and ensuring the public health by identifying pathogens in the environment. Many of these techniques have been made possible by a technology called the polymerase chain reaction (PCR). Before 1985, hardly anyone had heard of PCR; by today's standards, a laboratory without a PCR machine is like an office without a photocopier. In a sense, the PCR machine even functions like a photocopier: It takes a fragment of DNA and automatically turns out millions of copies in a scant three hours.

A second essential element in the DNA analyses is a fragment of DNA called the DNA probe. Developed in the 1970s, the DNA probe (or gene probe) hunts for a complementary fragment of DNA within a morass of cellular material and signals when the fragment has been located. Tracking down a gene or set of genes can be a formidable task when one considers the size of the human genome: There are over three billion nitrogenous bases in the forty-six human chromosomes. A gene probe attempts to detect a stretch that is a few thousand bases in length, or about a millionth of the length of the genome.

Here we will explore the fruits of PCR and DNA probe technologies as they apply to analyses and diagnoses. In many cases, these technologies are supplementing or replacing the traditional methods of analysis. Indeed, certain diagnoses cannot be made without the technologies involving DNA.

The PCR operates on the principle of target amplification, that is, the target DNA is amplified before the DNA probe is added. An alternative principle is signal amplification. According to this principle, a minuscule amount of target DNA binds to the DNA probe and then the signal is amplified to indicate that a successful match has been made.

To achieve signal amplification, there are two general approaches. The first is to decrease background "noise" by eliminating excessive DNA. This is accomplished by separating the target-probe complex from the mixture of contaminating DNA. For the separation, strands of thymidine nucleotides (poly-T) are linked to magnetic particles. Complementary strands of adenosine nucleotides (poly-A) are then linked to the DNA probe. After the DNA probe has had a chance to bind to its target DNA, the poly-T particles are added. Now the adenine bases of the target-probe complex bind to the thymine bases of the particles. The particles can then be separated from the mixture, carrying along the target-probe complex. So concentrated and separated from the contaminating DNA, the signal (the radioactive DNA probe) is amplified.

The second approach is to amplify a detector that binds to the signal, that is, the probe DNA molecule. To achieve this goal, a second probe for the target DNA is used. This probe is composed of RNA having a unique tertiary structure. The probe works with Q-beta replicase, an enzyme that catalyzes RNA replication. The RNA probe binds to the target DNA along with the probe DNA. Once the three have bound together, the RNA-DNA-DNA mixture is separated from the mixture and Q-beta replicase is added. The enzyme replicates the RNA probe geometrically (thus amplifying the detector [the RNA] and the signal [the probe]), and the large amount of RNA can be detected by standard techniques.

The RNA probe system is advantageous because it is very rapid, each cycle of RNA replication occurring in fifteen to twenty seconds. The technique also can be used to quantitate the target DNA and to detect RNA.

Just as the integrated circuit chip has revolutionized the personal computer industry, so too the new DNA microarray GeneChip probe array, may push the biotechnology industry a quantum leap forward. The GeneChip probe array is a glass slide the size of a postage stamp (about two centimeters square) that contains thousands of DNA probes, which scan thousands of DNA samples at one time. The chip reduces the time necessary for a DNA detection from days to hours.

Designed by scientists at Affymetrix Inc., the GeneChip probe array is built by allowing stencil-like masks to deposit various known DNA sequences (DNA probes) on a glass slide at sites called "features," which can be as small as twenty by twenty micrometers. The sequences then hang out from the glass slide. After thirty-two automated steps, a chip may contain more than 65,000 probes. The average length is about 20 bases.

To use the chip, an unknown DNA molecule is broken into fragments, and fluorescent marker molecules are attached to the fragments as signal devices. The fragments are pumped underneath the glass slide and forced over the probes; then the residue is washed away. Fragments finding a complementary sequence stick to the probes like zippers and deposit their fluorescent signals. A laser beam now scans the slide, row by row, and excites any fluorescent molecules present. A computer records the pattern of bright lights and indicates which probes united with which fragments. Smaller probe sequences are constructed to form larger sequences, and soon a long map of a genetic sequence emerges. Comparing the sequence to a map of known genes reveals the nature of the gene.

A DNA microarray could also detect a mutation in a gene. For example, if a gene has a base sequence of ...CCCAGGGG, it will bind at a feature where the probe contains the bases ...GGGTCCCC. But if the gene contains a single base mutation, its sequence might be ...CCCCGGGG, and it would bind to the feature where the probe has the sequence ...GGGGCCCC. The signal emitted at that feature would indicate the presence of the mutation.

As of 1998, Affymetrix was producing GeneChip probe arrays with more than 400,000 probes per chip. Although design is a lengthy process, production is automated and time-efficient, and the prospects for rapid gene probe analysis are positive. It is hoped that the DNA microarray may someday hold all the probes for the entire genome of an organism. By 1998, the GeneChip probe array was being used to screen for mutations in the p53 and BRCA1 genes, both associated with cancer and both discussed later in this chapter.

It may be unnecessary to identify a particular gene if a suitable alternative exists. Such an alternative is the restriction fragment length polymorphism, or RFLP (pronounced "rif-lip"). A RFLP is a stretch of DNA serving as a marker for a specified gene. The RFLP has no apparent function in the cell, and various RFLPs are located randomly throughout a person's chromosomes as a type of genetic litter. They are useful as marker genes when the particular gene has not yet been cloned or is difficult to work with.

Detecting a RFLP
The first individual has DNA with three sites sensitive to Hindlll, so that two fragments can be produced. The first fragment is 2 kilobases (kb) long; the second is 4 kb long. The second individual lacks one of the sensitive sites (H3 possibly because of an inherited mutation. Thus, when the Hindlll restriction enzyme is added, it only cuts the DNA molecule at two sites and produces a 6 kb fragment. Note that the DNA probe for the DNA segment between sites H and H'" has complementary bases for the entire segment, so it will react with the 2 kb, the 4 kb, and the 6 kb segments. The probe can therefore be used to detect all three segments. The segments are RFLPs.

For a RFLP to be used as a marker, it must be less than 5 x 106 base pairs (or 5 centimorgans) away from the gene. If the RFLP and gene are within this distance, there is a high degree of certainty that they are linked biochemically and move together from parent to offspring during reproduction. DNA technologists are statistically certain, with ninety-five percent confidence, that if they can locate the RFLP in an individual, the gene they are seeking will also be present.

To understand the nature of a RFLP, we must note that a particular stretch of DNA can be broken into fragments of various size. The fragments are called polymorphisms (literally translated, "many forms"). Consider, for example, the following stretch of DNA extending from points A to F and where the double letters indicate points of restriction enzyme cleavage:

-A-BB-CC-DD-EE-F

Let us now add a mixture of restriction enzymes (b, c, d, and e) that will cleave the DNA molecule at the double letters. The result will be five fragments of roughly equal size:

-A-B B-C C-D D-E E-F-

But over the great expanse of time, changes have occurred in DNA molecules: spontaneous changes in base pairs that have no bearing in biochemistry or physiology but do influence enzyme activity. Thus, a change in base pairs may have occurred in the stretch of DNA to produce the following molecule:

-A-BB-CC-DX-EE-F

If we now treat this molecule with the restriction enzyme mixture used before (b, c, d, and e), the following fragments will result:

-A-B B-C C-DX-E E-F

Note that the result of this enzyme activity is only four fragments because enzyme "d" is inactive at the DX site owing to the change from the original DD to the present DX. Also, and of practical significance, one fragment (C-DX-E) is longer than the other three fragments. Thus the stretch of DNA exists in fragments of various sizes; the derived from "restriction" enzymes and the fact that the length of the fragments can be determined; hence FL "fragment length.")

To use RFLPs in a DNA analysis, the DNA technologist takes advantage of the fact that the base-pair length of a particular RFLP is known. A person's cells are obtained and the DNA is isolated and treated with the restriction enzyme mixture. Then the mixture is placed in an electrophoresis apparatus (discussed in Chapter 9), and the polymorphisms are separated according to size. The separations yield a pattern of bands similar to a supermarket bar code. When the pattern is analyzed, the band reflecting the particular RFLP may or may not be present, depending on the inheritance pattern of the individual. If the marker RFLP is present, it may be assumed with ninety-five percent confidence (nineteen chances in twenty) that the gene is also present.

In practical terms, a particular chromosome and its genes can be followed from generation to generation by following the marker RFLP. When a disease-associated gene is analyzed, the RFLP pattern of an individual is compared with the RFLP patterns of normal relatives and relatives affected by the gene in question. Such comparisons make it possible to determine whether the individual has the marker RFLP and the disease gene. The information can then be used to make important decisions on how best to deal with the genetic disease in question, as we will see in the discussions to follow.