Legal and Ethical Considerations

The legal system, in both the criminal and civil arenas, may well be revolutionized by the advent of forensic DNA typing. One state trial judge has written that DNA typing “can constitute the single greatest advance in the ‘search for truth,’ and the goal of convicting the guilty and acquitting the innocent, since the advent of cross examination.” People v. Wesley, 140 Misc.2d 306, 533 N.Y.S.2d 643 (Co. Ct. 1988). A prominent professor in the field of law and forensic sciences believes that “DNA analysis will be to the end of the 20th century what fingerprinting was to the 19th.” The Washington state legislature has found the accuracy of DNA identification to be superior to that of any presently existing technique, and the Maryland General Assembly has proclaimed DNA identification to have been refined to a level of scientific accuracy that approaches an infinitesimal margin of error. Indeed, the forensic applications of DNA typing are limited only by the circumspection of the criminal mind. Regardless of the type of crime committed, whenever trace evidence appropriate for DNA analysis is left behind by the perpetrator and later recovered by the police, the test results can be an important investigative tool. Most frequently, such evidence will be found as a result of violent crimes. With 92,490 rapes and 20,680 homicides in the United States in 1988, the forensic application of DNA typing should significantly increase the arrest and conviction rates. The use of DNA typing is not confined to those cases in which body fluids, hairs, or tissue are left at the crime scene or on the victim by the perpetrator, thereby connecting a suspect to the scene or victim. Just as common is the situation where evidence is left by a victim on the suspect or the suspect’s belongings, which will establish previous contact between the accused and the victim. The Joseph Castro case is an example of this situation. He was accused in New York of stabbing to death a 20-year-old woman who was seven months pregnant and her 2-year-old daughter.

Laboratory Organization

Laboratory organization involves both the physical establishment and its operation. It is perhaps simplest to divide the laboratory into its component sections and discuss each separately. The areas may physically overlap for a small facility, and depending on the operation specialty, some sections, such as tissue culture or probe amplification, may not be required. Unless the operation is large, dark room and cold room facilities and expensive equipment such as an ultracentrifuge and beta counter are best shared, if feasible. The setup of a DNA analysis facility is a relatively simple process if it is incorporated into an established biochemistry program; it is considerably more involved if no such base exists. The outline presented in this chapter is only a guide; individuals contemplating the development of a new facility should visit as many established centers as possible. Discussions with sales representatives and attendance at relevant trade shows and DNA conferences are invaluable. Office requirements for a DNA program are no different in principle from those of any other biochemistry program. At least one separate office is required, usually for the program director and, as space permits, offices for a clerk-secretary and senior technologist are useful. Everyone working in the laboratory must have at least a small partitioned desk space in a quiet location. Lockable fire-resistant cabinets are required to store sensitive records; these cabinets should be accessible, preferably located in the clerical area. Analysis results are worthless without proper documentation of a specimen’s chain of custody (continuity). Information, including time and conditions of specimen procurement, conditions of storage and shipment, date received by the laboratory, and reason for the analysis request is also required. These data can be manually recorded; however, entry into a computer program capable of sorting and maintaining records for long-term retrieval is almost mandatory. Storage of unprocessed specimens may be necessary, and if at all possible, DNA should be isolated when received.

Probability and Statistical Analysis

Determination of the probability of specimen match and estimation of population allele frequency distributions are two key areas of DNA profiling requiring probabilistic and statistical analyses. Statistical calculations can be tedious and slight changes in the wording of probability statements can result in vastly different meanings. It may, therefore, be prudent for the analyst to seek the advice of a qualified statistician before assembling population frequency data or submitting probability statements in a court of law. Statistical methods provide powerful tools to assist with decision-making. Because of easy access to powerful statistical software on personal computers, these methods are easy to use. However, for this same reason they can also often be misused and questionable data presented in a favorable light. One must always be aware of those skilled in the misleading use of statistics or those who simply make incorrect statistical statements even though the quantity of base data may be considerable and the quality good. The objective of this chapter is to review methods of probability and statistics relevant to DNA fingerprint analysis. Basic statistical data are usually derived from samples drawn from large populations. A population is a collection of individuals having stated features in common, such as all Orientals in the United States. A simple random sample is a subset of individuals selected from a population using a random choice mechanism (such as a random number table) which guarantees that all members of the population are equally likely to be chosen. Usually the sample size is denoted by n. Values in a sample are called observations and denoted by X1, X2, . .. , Xn. The features of interest in a population such as the true average IQ, or the true proportion of a specific allelle at a given locus, are called parameters, while statistics are numerical summaries of data such as means or standard deviations. The science of statistics is concerned with inferring information about population parameters based on sample statistics.

Analysis Techniques

Conventional DNA analysis techniques include cleavage of DNA by restriction enzymes, fragment electrophoresis, Southern transfer, probe labeling, probegenomic fragment hybridization, and print detection (Cawood 1989, Sambrook 1989, Berger 1987). Details of the assay conditions may vary considerably depending on the specific probes hybridized. Endonuclease digestion, electrophoresis, and Southern transfer are not required with simple dot-blot procedures. The quality of the final result can be no greater than the quality of the input DNA specimen and the attention of the analyst to assay details. The format of the analysis blot must be carefully considered to include control specimens and a broad range of size markers. The analyst must also be certain about the sizes of the profile fragments to accurately determine if matches exist between crime evidence and suspect specimen or offspring and putative parent specimens and to calculate the match probabilities. Restriction enzymes cleave DNA at specific recognition base sequences. It is important to choose an enzyme with sites flanking the repeats when fragments consisting of different numbers of tandem repeats are to be characterized for DNA profiling. Cleavage within a repeat sequence will result in the production of small fragments that may be unresolvable. The choice of enzyme, in this respect, is accomplished either by trial and error or by knowledge of the base sequence of the fragment flanking regions. The optimum reaction conditions vary for each enzyme; consequently, suppliers usually provide information sheets for the user. Digestion temperature and buffer salt concentration are the critical features. The reaction mixture can be overlaid with a few drops of paraffin oil to prevent vapor formation and changes in the buffer concentration. This applies mainly to enzymes such as Taq I that require high reaction temperatures (65°C in this example). Unless specifically indicated otherwise, three different strength ionic buffers will accommodate most enzymes.

Genetic Principles

Genetics is the study of heredity. Each individual’s makeup, or phenotype, is determined by nature and modified by environmental factors. DNA identity analysis is based strictly on heredity, and only in the rare case where a human had a bone marrow transplant would the white blood cell genotype differ from that inherited. Difficulties can arise with specimens because of DNA degradation or contamination by extraneous materials, and mixed cell populations could be present in tumorous tissue. The analyst must always be cognizant of these complicating factors. The concept of the gene was advanced by the Moravian monk Gregor Mendel in 1865 based on observations he made after crossing different varieties of garden peas; these experiments are considered the beginning of the discipline of genetics. (The term gene was actually coined by the Danish plant scientist W. Johannsen in the early 1900s.) Mendel formulated two laws. The law of segregation or separation states that two members of each gene pair (alleles) in a diploid organism separate to different gametes during sex cell formation. The law of independent assortment states that members of different pairs of alleles, if located on separate chromosomes or far apart on the same homologous chromosome pair, assort independently into gametes. These laws are basic to the understanding of biological family relationships and play a critical role in such contemporary issues as paternity testing and immigration disputes. The basic unit of life is the cell. Cells are microfactories in which raw materials (amino acids, simple carbohydrates, lipids, and trace elements) are received, new substances (proteins, complex lipids, carbohydrates, and nucleic acids) are produced, and wastes are removed. The thousands of different enzymes required for the myriad ongoing chemical reactions are key to the efficient functioning of cells. Each cell has the ability to self-replicate using the deoxyribonucleic acid (DNA) code as the blueprint, raw materials as building blocks, and enzymes as catalysts. It has been estimated that the average human being is composed of approximately 100 trillion cells—a considerable amount of DNA.

Case Applications

A number of forensic and family relationship cases, as well as medical, animal science, wildlife poaching investigation, and plant science applications are presented in this chapter. As suggested by the titles and headlines from various journal and newspaper articles, the process of identification using recombinant DNA technology has proven to be very practical. Semen from a rape case, a hair follicle from a homicide, blood stains at a break-in, chorionic villi from a prenatal diagnosis, blood cells from a transplant patient, tumorous tissue, a big game animal gut pile, a freezer steak, rare condor blood, a whale skin biopsy, plant tissue, and ancient human and other animal remains are some of the sources of DNA used for typing. Perhaps the most apparent indicator of application potential can be deduced from the number of recent patent applications covering recombinant DNA processes and products. In addition, many new government and commercial ventures have been established to accommodate the anticipated service load. The analysis of DNA is providing hard evidence for the resolution of serious criminal acts and other difficult identification problems in homicide, rape, accident, missing persons, break-ins, and hit-and-run cases (Anderson 1989; Barinaga 1989; Conner 1988; Dodd 1985; Fowler 1988; Fox 1989; Fukushima 1988; Gill 1987; Giusti 1986; Hicks 1989; Higuchi 1988; Hewlett 1989; Jeffreys 1988; King 1989; Kobayashi 1988; Lander 1989a; Lewin 1989; McElfresh 1989; Marx 1988; Merz 1988; Newmark 1987, 1987a; Norman 1989; Ross 1989; Taylor 1989; Yokoi 1989). The determination of whether a series of crimes is serial or copycat, that is, committed by one or more than one perpetrator, is critical to the investigation of many cases. If DNA profiles match for specimens from different crime sites, this suggests that the same individual was involved and investigators can then concentrate their efforts on the hunt for one person. The forensic scientist first prepares a DNA identity profile of the crime (evidence), suspect, and victim specimens.

Probes, Allele Mutations, and Restriction Enzymes

Positive identification is the ultimate objective of forensic analysis of blood and other tissue specimens. Nucleotide probes can be very effective tools for detecting genetic markers in this identification process. The genetic markers should be highly polymorphic; allelic variants should be easily and readily detectable; if amplification is required, the alleles should be efficiently amplified using PCR technology; and a statistically sound estimate of the population allele and genotype frequencies should be available. Probes are single-stranded fragments of DNA or RNA containing the complementary code for a specific sequence of genome bases. Probes available for DNA profile analysis will, no doubt, eventually number in the hundreds. Currently, the most valuable detect tandem repetitive sequence fragments either at a specific locus under high-stringency analysis conditions or at numerous loci under low-stringency conditions. Each locus consists of many possible alleles with frequencies that vary depending on the specific population. Other factors also enter into the selection of probes, including ease of amplification, stability, cross-reactivity, and general availability. Rate of allele mutation is also a prime consideration in probe selection. Mutation can be considered at two levels: as the basis for the large number of tandem repeat (VNTR) alleles formed during evolution and as a possible reason for spurious unassignable bands in typing analysis. Although highly unlikely, somatic mutations may be of concern in forensic testing if DNA from different tissues, such as blood and hair roots, are being matched. Germ line (gamete) mutations must be considered when parentage analyses are undertaken. These situations could give rise to false negative results and, therefore, false exclusions. Different considerations also apply for single versus multilocus probes. If a band that is not seen in the putative father is detected in an offspring, the man could incorrectly be excluded if the single-locus probe approach is used. This situation would necessitate testing with more than the usual four or five probes.

DMA Amplification

Amplification of DNA may be necessary to increase the quantity of sample available for profiling, to reduce the analysis time, or to produce probes for the hybridization process (Higuchi 1989, Li 1988, Marx 1988, Mullis 1990, Paabo 1989, Saiki 1986). Stretches of nucleotides up to at least 3,000 bp from any DNA-containing samples may be efficiently amplified by the polymerase chain reaction (PCR). Alternatively, living tissue can be placed in culture, and fibroblasts, epithelial type cells, or lymphoblasts grown. The culture process differs considerably from the PCR approach in that the total genome is reproduced. Also, tissue culture is usually at least a two-week procedure, whereas the polymerase chain reaction requires only a few hours. Cultured cells can be used for enzyme and other biochemical tests, and storage in liquid nitrogen is a standard practice for regrowth at a later time. Probe material, that is, DNA capable of hybridizing with its complementary region in the genome, must be amplified, aliquoted, and stored to provide an ongoing source for use with each profile analysis. Probe amplification has been mainly carried out in bacterial culture; however, probes can be chemically synthesized as discussed in Chapter 2 or amplified by the PCR system. At least 10 to 50 ng of high molecular weight genomic DNA are required for VNTR analysis using single-locus probes, and at least 0.5 to 1.0 μg required if multilocus probes are used. If only a small quantity of DNA is available, amplification using the PCR may be the only feasible option for obtaining sufficient material for analysis. PCR has revolutionized the approach to the recovery of DNA from a variety of sources. Microgram quantities of DNA can be produced in vitro by the amplification of picogram starting amounts. Single-copy genomic sequences greater than 2 kb in length have been amplified more than 10 millionfold in a few hours. Amplified material can also be directly sequenced without the necessity of incorporating DNA fragments into vectors such as M13 (Gyllensten 1989,1989a). Availability of oligonucleotide primers is the key to the amplification process.

Specimens

The proper handling of specimens for direct storage or DNA extraction and characterization is one of the most important aspects of the profiling procedure. Because DNA typing is not yet a routine test, some laboratories may perform only the isolation portion of the overall analysis and leave the other methodologies to specialized centers. Profiling may never be required for many forensic specimens and only intermediate storage needed. It is essential that smaller centers have at least the facilities to isolate, characterize, and store DNA. A broad range of DNA sources exists. Fresh tissue usually includes whole blood, buccal epithelial cells, and hair follicles. Under special circumstances, in the medical setting, the genotyping of amniotic fluid cells, chorionic villus samples, and tissue culture fibroblasts may be required. Dried specimens usually include blood and semen stains, tooth pulp, and bone marrow. Animal trophy heads and pelts are also sources of dried DNA. Preserved or unpreserved human autopsy specimens, and tissues from animal gut piles and frozen meat, are other possible sources of DNA. As with any biological test, the quality of the results can be no better than the quality of the input sample. If the DNA is highly degraded or contaminated, it may be unusable; thus, every effort should be taken to collect, record, transport, store, and isolate materials using meticulous techniques. The specimen of choice is 1 ml or more of fresh whole EDTA blood. Anticoagulants other than EDTA may be acceptable; however, there are reports that heparin interferes with the activity of certain restriction enzymes. The quantity of DNA isolated from 1 ml of blood is usually sufficient for the necessary testing and a considerably smaller sample will often suffice. There are occasions, however, when the DNA yield is low and a repeat specimen is required; for this reason it is prudent to collect an additional sample if possible. Buccal epithelial cells obtained from mouth swabs, and hair follicles are two other general sources of fresh DNA; however, the DNA may require amplification by the polymerase chain reaction to provide sufficient material for analysis.

Introduction

DNA identification analysis, identity testing, profiling, fingerprinting, typing, or genotyping refers to the characterization of one or more relatively rare features of an individuals’s genome or hereditary makeup. Every human, lower animal, and sexually reproduced plant has a characteristic phenotype or physical appearance because each possesses a unique hereditary composition. The exception to this rule is identical twins, who possess the same unique genotype but, owing to the consequences of complex developmental events, have subtly different phenotypes. The DNA of any individual is identical whether it is extracted from hair bulbs, white blood cells, or a semen specimen. These principles of individual uniqueness and identical DNA structure within all tissues of the same body provide the basis for DNA profiling. Identity testing is only one aspect of recombinant DNA analysis. It is the newest, most powerful technique in forensic science, paternity testing, animal and plant sciences, and investigation of wildlife poaching. The foundation for the concept was established with the hallmark observation by Wyman and White (1980) of a polymorphic DNA locus characterized by a number of variable-length restriction fragments called restriction fragment length polymorphisms (RFLPs). The history of DNA fingerprinting, as such, is even more recent, dating from 1985 with the paper “Hypervariable Minisatellite Regions in Human DNA” by Alex Jeffreys etal. Jeffreys and his coworkers were analyzing the human myoglobin gene when they discovered a region consisting of a 33-base-pair (bp) sequence repeated four times within an intervening sequence (IVS). This tandem repeat was referred to as a minisatellite and similar regions as being hypervariable because the number of tandem repeats is variable both within a locus and between loci. They also discovered that each repeat unit contains a smaller 16-bp core in common with other minisatellites. When DNA is isolated, cleaved with a specific enzyme, and hybridized under low-stringency conditions with a probe consisting of the core repeat, a complex ladder of DNA fragments is detected. This profile appears to be unique to each individual. Different core repeats were later isolated and used to produce a number of different probes useful for fingerprinting.