The successful prosecution of Edward Beamish depended on the willingness of the Supreme Court of Prince Edward Island to admit evidence of identity that was based on the use in the laboratory of the polymerase chain reaction. PCR was conceived by Kary Mullis, a young, unconventional, molecular biologist, while he was working for Cetus, a biotechnology company in California during the early 1980s. In essence, Mullis realized that there was an astoundingly easy way to use well-characterized and commercially available enzymes to amplify any sample of DNA, however small, until one had as much of it as one needed. This discovery, which garnered Mullis a Nobel Prize in 1992, has revolutionized molecular biology. It has ushered in a new era in cancer diagnostics and in paternity testing, greatly accelerated our efforts to find and clone human genes, and added an important new method to study speciation and biological diversity, to name just a few of its impacts.

PCR involves a sequential series of reactions, repeated over and over. First, the DNA of interest is heated, which causes its two strands to fall apart. Next, very short stretches of single-stranded DNA of known composition are chemically connected to each end of the target DNA. If you could see it, the DNA would now look double-stranded at one end, single-stranded for most of its length, and double-stranded at the other end. The last step is to add the four basic DNA building blocks and an enzyme called DNA polymerase to the mixture. The polymerase recognizes the end of the double strand and chemically adds a new complementary strand to it opposite the middle single strand, continuing until it reaches the other double-stranded region. At the end of the first cycle (which takes minutes), the amount of the original DNA has doubled. As one repeats the series of steps, one keeps doubling the copies of the target DNA. A series of 20 runs amplifies the original amount 1,000,000 fold.

From a forensic perspective, PCR greatly extends the detective's ability to find usable evidentiary samples at a crime scene. It is almost magical in its ability to amplify DNA, a fact that was understood very early in its development. As far back as 1988 in one of the earliest publications pointing out its possible uses in forensics, a scientist demonstrated that he had been able to use PCR to extract and characterize DNA from a single hair root, thus anticipating the work with Snowball. In 1989, scientists showed that they could retrieve small bits of mitochondrial DNA from a 5500-year-old-skeleton and amplify it with PCR.

In 1991, Alec Jeffreys, the British scientist who in 1985 had launched DNA forensics using a different kind of technology, became the first to use PCR analysis of bone DNA to positively identify a murder victim. The badly decomposed remains of a 15-year-old girl had been found wrapped in a carpet in a shallow grave. From study of the skull and dental records, forensic anthropologists had tentatively identified the remains as those of a teenager missing since 1981. Jeffreys extracted the DNA from a small sample taken from the interior of the thigh bone. About 90% of it was actually DNA from soil bacteria, and the human DNA was badly degraded into very short stretches. He used short CA repeats at six different loci and was able to type the DNA. He was then able to compare it to DNA taken from the victim's presumptive parents and confirm the identity, despite the weakness of then-available reference population data. A year later, Dr. Mary-Claire King, a Berkeley scientist who was soon to gain fame for proving the existence of a gene in which mutations predisposed to hereditary breast cancer, became the first to use PCR to establish the identity of a murder victim by analyzing the DNA in teeth.

These first reports created a major stir in the forensics and biotechnology communities. In 1993, Cetus, the company at which Mullis and his colleagues had developed PCR, began marketing a PCR-based kit for studying forensic samples. Within a year many prosecutors were having their first go at using PCR-generated evidence in court.

Especially with regard to criminal law, the courts are conservative in evaluating proposals to admit evidence based on new scientific methods. The courtroom floor is littered with technologies, such as early "lie detector" tests and the paraffin analysis of gun shots, that did not stand up to scrutiny. The justices on the Supreme Court of Prince Edward Island were willing to admit PCR-generated evidence only because by 1996 a sufficient number of other judges around the world had already scrutinized similar evidence, satisfied themselves that there was a firm scientific basis to admit it, and published their opinions.

A murder trial in Massachusetts at which I testified illustrates the power of forensic PCR. In the summer of 1994, a man was bludgeoned to death in a woodshed in the town of Athol, a rundown city located midway between Boston and the Berkshires. The bloodstained murder weapon, a 4-foot-long 2 by 4, lay nearby. Detectives sealed off and painstakingly studied the crime scene. Among the many items they picked up for analysis were several cigarette butts. Forensic analysis found that all the blood at the crime scene derived from the victim. The laboratory next decided to use PCR technology to attempt to characterize DNA from the dried cigarette butts. After using sophisticated methods to extract DNA from the few cells that might be caught in the fibers of filter paper, the laboratory technicians then used a relatively new system of forensic probes developed by Cetus to characterize the DNA. They succeeded in constructing a DNA fingerprint. But the evidence was of value only if it could be linked to a suspect.

The police had a single lead based on a sketchy eyewitness report. On the night of the murder, someone saw a man, whom he could describe moderately well, casually pushing the victim in his wheelchair along a street near the crime scene. The report suggested that the killer knew his victim, which is often the case, and gave the police the rationale they needed to focus their investigation on the whereabouts that night of just a few persons. They eventually arrested a young man, charged him with the crime, and obtained a court order to obtain his blood for DNA analysis. Forensic analysis matched his DNA with the results of the cigarette butt studies. However, the prosecution still had a problem. The suspect, who acknowledged knowing the victim, claimed that he had been at the crime scene many times in the past, and had dropped cigarettes there on many occasions. He did not have a strong alibi for his whereabouts at the time of the murder.

I became involved in the case at the eleventh hour. Late one afternoon the court-appointed defense attorney, who had been searching for weeks for an expert to challenge the use of DNA evidence generated by PCR amplification, got to the end of his list and called me. I was extremely reluctant. Because I am a clinical geneticist and had served on the national committee that issued the first major report on DNA forensics (one that barely mentioned PCR, which was then still in its infancy), I knew that I would qualify as an expert. But I had little laboratory experience with DNA testing and I had always avoided involvement in criminal trials. However, the defense attorney, an engaging former high school teacher named Harry Miles, whom I came to respect tremendously for his courtroom skills, won me over with an argument that I could not refute. He had no more names, the trial was to start the following week, and without an expert to at least raise some doubts about the PCR evidence, he could not provide his client with a full defense. He passionately believed that our criminal justice system could not function fairly if he could not produce someone to raise questions about PCR.

Thus, I found myself driving along route 2 through the beautiful little towns of north central Massachusetts, to a picture-postcard New England courthouse, rehearsing my arguments against a technology I essentially believed in. Miles wanted me available to rebut at least some of the prosecution's expert testimony. There was no small amount of irony here. The prosecution's expert was a young molecular biologist who had been working at the state forensic lab in Connecticut for about a year. His boss, Dr. Henry Lee, had sent him to testify in support of DNA evidence. Dr. Lee, a celebrated forensic expert who had worked on the national DNA forensics committee with me, was unavailable because he was in California at the O.J. Simpson trial testifying against the use of DNA evidence! At that time, the fall of 1995, the interminable Simpson case had been considering DNA evidence for weeks.

Miles knew that he was asking someone who generally believed that DNA evidence was accurate and trustworthy to testify against it. I was will ing to address only certain issues. In 1995 there were still some unresolved technical problems, mostly concerning the adequacy of the reference populations used to calculate the odds that DNA from a randomly selected (innocent) person would match the crime scene sample. I was comfortable discussing those. Furthermore, no forensic lab yet had much courtroom experience with PCR. The prosecution was using it for the first time, the defense was fighting it for the first time, and a Massachusetts judge was evaluating it for the first time.

As it turned out, the evidence in question was so novel that even my small attack on its validity was enough to help make the assistant district attorney tack in the direction that attorney Miles wanted. Before trial she had steadfastly refused to plea-bargain away from a murder indictment, and the defendant had resolutely proclaimed his innocence. A few days after the expert testimony, however, she accepted the defendant's offer to plead guilty to the lesser charge of manslaughter.

Since the early days of courtroom PCR, evidence generated by it has been used in several thousand trials in the United States. Only a handful of courts have refused to admit PCR and those almost always due to concerns about how a particular forensic lab did or recorded its analysis. In 1997 the Supreme Judicial Court of Massachusetts, a court that took a very cautious approach to forensic DNA evidence in the late 1980s and early 1990s, ruled that PCR-based evidence was admissible. Since then, courts in many other states have followed suit.

As scientists have become more adept at using PCR to amplify minuscule amounts of DNA, they have greatly extended its original application. In 1997, Roland Van Oorschot, a DNA forensic geneticist in Australia, really pushed the envelope. A knife that police thought had been used in an attempted murder was given to his lab for analysis. Although it did not have any blood on it, Van Oorschot reasoned that someone must have held it, so he swabbed the knife down and ran the standard PCR protocol. He came up with sufficient DNA to construct a DNA profile. This led him to hypothesize that people leave sufficient DNA on virtually whatever they touch to permit a lab to develop an identification profile.

To test the hypothesis, Van Oorschot and his colleagues tried to obtain DNA from common objects. Using cotton cloth moistened with sterile water and handled with sterile forceps, they swabbed and studied briefcase handles, pens, a car key, a mug, a glass, a telephone handset, gloves, and the insides of four condoms which had been worn but into which there had been no ejaculation (a study that is relevant to the investigation of a significant fraction of rapes). They also swabbed hands. Virtually all the objects yielded enough DNA to identify the user. The scientists even swabbed the hands of men whom had recently shaken the hands of other persons. They showed that merely by shaking hands there is a transfer of DNA. That is, under controlled circumstances they were able to identify who had most recently shaken hands with the person whose hand was swabbed and studied. They also showed that it is possible to obtain DNA from the fingerprints left on glass.

The Australian detectives and colleagues in many nations are now taking a new approach to crime scene investigation. They are assuming that every criminal leaves DNA at the scene. When evidence is gathered properly, the yield has been astounding. For example, forensic scientists have identified DNA obtained from inside a glove left at a burglary, from the rim of a glass, and even from a fingerprint on a window sill. Detectives in England have used DNA to revolutionize the approach to solving auto theft. They now routinely swab the steering wheels of recovered stolen vehicles. DNA analysis typically yields two or more profiles. Of course, one almost always belongs to the owner. The second profile is often that of the thief who drove the car for a while and then abandoned it. Oddly enough, no one is quite sure where the DNA comes from. The best guess is that it comes from keratinized epithelial cells that break open on the surface of the skin, but these cells have very tough coats, so scientists are dubious. Some think that naked DNA may actually be extruded through sweat glands.

The fact that it may be virtually impossible to be involved with a crime without somehow leaving DNA on some object connected with it triggers immensely challenging questions about the right of the state to use DNA analysis to solve crimes and the right of the citizen to maintain his genetic privacy. Our society may decide that the potential gains in crime prevention constitute an overwhelming argument in favor of mandatory universal DNA profiling at birth. If people shed DNA wherever they go and if there is a routine way to identify crime scene DNA by running the profile against a population-wide directory, it will almost always be possible to identify a person the first time he commits a felony, an extremely important development, given the number of serial criminals. DNA analysis of crime scene evidence will often find biological samples of innocent people, but, in most cases, it should be possible quickly to eliminate them as suspects. Merely showing that one's DNA was found at a crime scene or in connection with it would rarely be enough evidence to convict, but it could often provide strong circumstantial support to the prosecution's case.

Universal DNA banking seems inevitable. Its value in assisting the resolution of virtually any kind of serious crime will, I think, prove irresistible. The speed with which European nations and the various states in the United States have initiated DNA databanks on convicted felons is a harbinger of the future. One can anticipate a sharp debate over whether mandatory DNA sampling for identification purposes violates the constitutional right to privacy, but, in the end, all our DNA profiles will be in a computerized database. In the meantime, there appears to be little reason to wear gloves when committing a burglary; one's DNA is being shed from other bodily surfaces even as one moves.

Illustration of the amplification of DNA by the polymerase chain reaction (PCR).

(Reprinted, with permission, from Rosenthal 1994 © Massachusetts Medical Society. All rights reserved.)

0 0

Post a comment