Student Work
Bianka Bommarito and Samantha Hilber, Biological Sciences students, presented their honors thesis research in the student competition at the annual meeting of the American Society of Ichthyologists and Herpetologists, in Tampa, Fla., July 6, 2005. Bommarito's oral paper was entitled "Choice of Spawning Temperature by Neotropical Cichlid Fishes." Hilber's paper was titled "Egg Size Through Consecutive Spawning in the Convict Cichlid."
Ameer Thompson, biological sciences student, presented a poster at the National 2003 Sigma Xi Research Conference Nov. 14 in Los Angeles. It was titled "A Test of the Misdirection Hypothesis for False Eyespots Using Parental Convict Cichlids," and was rated "superior" by the panel of judges.
UNDER THE DIRECTION OF RUTH ELIZABETH BALLARD
Some of the work performed by our students involve the following research:
Her research program focuses on DNA FORENSICS - the the use of DNA to solve crimes. Currently, there are three major projects in the lab:
- Development of a system of population-specific human alleles for use in forensic casework
- Nuclear STR typing of DNA from shed (telogen) hair (a collaborative project with the California Dept. of Justice)
- Determination of the frequencies of nine STR loci among the Masaai of Northern Tanzania and the Hispanics of California's Central Valley.
1. POPULATION-SPECIFIC HUMAN ALLELES.
( Graduate Students : Mary Hansen, Betsy Garfield; Undergraduate Assistants: Adam Hagen, Kristi Goulet)
When forensic scientists prepare and analyze DNA samples collected at a crime scene, their ultimate goal is to obtain information that will eventually help prosecutors convict guilty individuals and help defense attorneys clear the innocent. Successful analysis currently involves following rigorous guidelines for the extraction of DNA from small amounts of starting material (e.g. blood, semen, strands of hair) and testing the sequence of the DNA to provide a DNA profile of the person who inadvertently deposited the sample at the crime scene. Generally, the testing is performed by a very sensitive procedure called the polymerase chain reaction (PCR) and the profile is based on DNA sequence differences between individuals at specific sites in the human genome called polymorphic sites.
The polymorphic sites (also called markers) commonly used in DNA analysis include VNTRs (variable number tandem repeats), regions of genes coding for blood surface antigens, and STRs (short tandem repeats). The DNA sequences are variable for one of two reasons. In the case of VNTRs and STRs, the DNA is intergenic (located in regions between genes) and thus is not under the same selective pressure as sequences that code for proteins. In the case of the blood surface antigen-coding genes, the genes themselves have been selected for variability to enhance an organism's ability to distinguish self from non-self. Thus, there is a high level of heterozygosity at these sites in the human population (heterozygotes are persons that carry two different forms of the same DNA sequence) and any particular sequence (allele) is found at low frequency. Determining the alleles carried by an individual, then, is useful a way of distinguishing that individual from other individuals in the population. For example, if allele 1 is found in 0.1% of the population and allele 2 is found in 0.2% of the population, the probability that a person will have both of these alleles is 0.02% - i.e. only one person in 5,000 would be expected to carry this combination. By performing this type of analysis using the polymorphic markers currently in use, forensic DNA scientists can exclude most of the earth's population as being a possible source of a sample.
One important limitation of this type of analysis is that it reveals almost no information about the physical characteristics (phenotype) of the individual who left the sample at the crime scene. Given a battery of known suspects, current DNA techniques can eliminate all but the guilty party and provide powerful statistical support for convicting that individual. However, the techniques cannot help law enforcement officers narrow down their search for suspects other than by sex. The presence of a Y chromosome in males allows rapid sex typing of crime scene samples but other physical characteristics such as ethnicity, hair color, eye color, and height remain beyond the scope of the current technology. Eventually, careful analysis of the sequence data from the Human Genome Project will result in the elucidation of the contributions of various genes to human traits and a system that provides phenotypic information will be developed. However, at the present time forensic DNA scientists are decades from this goal.
In lieu of waiting until the Human Genome Project is complete to begin developing DNA marker systems that reveal suspect phenotypes, our laboratory is currently developing a novel approach that exploits individual differences in human Alu sequences. Alus are highly repetitive, mobile DNA sequences that are widely interspersed throughout the human genome. They multiply by retrotransposition, a process which involves transcription of the Alu into RNA, reverse transcription of the Alu RNA into DNA, and the integration of the "daughter Alu" into a new genomic site. While most of the Alus became fixed in specific genomic sites before the divergence of humans and chimpanzees, several Alu subfamilies are still actively retrotransposed. While the DNA sequences in this marker system, like those of all other systems currently in use, do not directly influence observable phenotype, there is evidence to suggest that the frequencies of Alu alleles differ significantly among human populations. Thus, they have the potential to allow forensic DNA scientists to identify the racial origin of a crime scene sample and, by extension, to provide valuable phenotypic information about the perpetrator of the crime as well. For example, if a crime scene sample can unambiguously be identified as Asian in origin, several phenotypes, including hair and eye color, can be assumed. More importantly, a whole range of suspects can quickly be eliminated, allowing law enforcement officers to more efficiently allocate their time, energy, and resources.
The most accurate marker systems currently in use, the VNTR and STR markers, exhibit more diversity within racial sub-populations than between racial groups themselves. The reason for this is that these markers radiated relatively recently, after the racial divergence of humans was mostly complete some 10,000 years ago. Thus, most VNTR and STR allele frequencies differ by less than ten percent between even distantly related racial populations. What is needed for racial identification is a system of markers that became fixed in human populations around the time that human population divergence was complete but that have radiated only slowly since that time.
One active family of Alus, sb-2, is particularly promising in this regard because it was retrotransposing most actively between about 100,000 and 10,000 years ago The rate of retrotransposition has apparently slowed since then as revealed by the fact that at least two sb-2 Alus appear to be present at much higher frequencies in some racial populations than in others (2, 3). For example, Milewicz, et al. showed that a sb-2 Alu located in intron 8 of the human type III collagen gene (COL3A1) is found in 35% of African Americans but in only 1% of American Caucasians. Thus, sb-2 Alus are ideal candidates for development of a DNA marker system that will allow racial identification of forensic samples of unknown origin. To date, more than one thousand of these sequences have been cloned and entered in the Genbank database but only a few have been analyzed for their relative frequency in different human sub-populations.
2. STR TYPING OF DNA FROM SHED (TELOGEN) HAIR.
( Graduate Student: Peter Sclafani)
Hair is the most common form of biological evidence found at crime scenes yet is the most underutilized for forensic DNA typing (1). The basis for this apparent paradox is that hair is easily and inadvertently shed but contains very few cells unless it is pulled out from the root. Therefore, shed hair usually lacks sufficient DNA for identity typing by standard methodologies, including the very sensitive polymerase chain reaction (PCR). While plucked hair that contains an intact root shaft may contain as much as 200 nanograms (ng) of nuclear DNA, hair shafts are limited to less than 10 ng (2). In addition, hair pigments can interfere with the PCR reaction, a problem that becomes especially acute when only trace amounts of DNA are available prior to DNA extraction (3).
To address this problem, researchers have exploited the fact that cells contain mitochondrial DNA as well as nuclear DNA. Since any given cell contains tens to hundreds of mitochondria, mitochondrial DNA is more prevalent than nuclear DNA. Yoshi, et al. demonstrated that regions of the human mitochondrial genome could be amplified from hair shafts that lacked an intact root (4). However, this approach has two important limitations that seriously undermine its utility for DNA identity testing. First, the mitochondrial genomes within a cell and/or cell population are often heterogeneous, leading to ambivalent typing results. Second, the highly informative polymorphisms common in the nuclear genome are not present in the mitochondrial genome. Thus, the level of discrimination produced by mitochondrial DNA typing is far below that of nuclear DNA typing and labor-intensive direct sequencing must be used to detect minor sequence differences.
We propose to solve this problem by developing a reliable method for forensic typing of nuclear DNA from shed hair. Successful completion of this objective will result in a standardized set of protocols that can be used by forensic DNA technicians to reproducibly obtain nuclear DNA profiles from shed hairs that do not contain intact roots.
In situ PCR is a relatively new approach for detecting low-abundance nucleic acid targets (for a review, see 5). The method involves fixing intact cells on Teflon-coated slides and performing the PCR reaction directly on the slide using a specialized thermocycler. The PCR reaction is then followed by in situ hybridization to detect the amplified DNA product. Although this technique has primarily been used to compare genotype and/or gene expression profiles across cell populations (5), researchers have also had success amplifying very low abundance DNA targets such as HIV-1 DNA in human brain tissue (6) and Epstein-Barr virus DNA in oral squamous cell carcinomas (7) that could not be amplified by standard solution PCR. Presumably, the sensitivity of the technique results from eliminating the need for DNA extraction and thus maximizing the amount of target DNA available for PCR. Thus, in situ PCR offers an excellent alternative to standard PCR when target DNA levels are unusually low. Because this is precisely the problem forensic technicians face when attempting to produce DNA profiles from shed hair, we propose adapting this technology toward completing the objective of this proposal.
Several DNA marker systems have been developed to allow identity testing in humans. Until recently, PCR amplification and detection of dimorphisms and polymorphisms at the DQA1 and "polymarker" loci was the method of choice because the detection technology was fast, sensitive, and reliable. However, the DQA1/polymarker system has now been replaced by a system of length polymorphisms called short tandem repeats (STRs) that provide a much higher level of discrimination. STR alleles are distinguished by amplification across the STR region followed by capillary gel electrophoresis (CGE) using a fluorescent detection technology. Our adaptation of the in situ PCR procedure will involve in situ PCR of the STR alleles, recovery of the amplified products by capillary withdrawal of the amplification fluid, re-amplification of the products in a solution PCR reaction, and fragment analysis using standard CGE protocols and data analysis software.
This project is a joint venture between Dr. Ruth E. Ballard of California State University, Sacramento and Theresa Spear, M.A., of the California Criminalistics Institute (CCI), a training facility for the California Department of Justice. The venture is perceived by both parties as beneficial to the objectives and on-going programs of both organizations. Dr. Ballard's research program will benefit from the resources and case-based intellectual input from Theresa Spear and CCI while Ms. Spear and CCI will benefit from the development of protocols that allow typing of shed hair in forensic casework. CCI has agreed to provide Dr. Ballard and her students with (1) access to the CCI library to obtain background case information, (2) training on and use of a PE Biosystems 310 Genetic Analyzer located at CCI, (3) bench space at CCI, and (4) aid in data evaluation and interpretation. Dr. Ballard will provide the equipment and reagents necessary for performing the microtome and in situ PCR steps, the student personnel to perform the bench work, and oversight of the design and development of the project at all phases.
1. FREQUENCIES OF NINE STR LOCI AMONG THE MASAAI AND THE HISPANIC POPULATION OF THE CENTRAL VALLEY OF CALIFORNIA.
( Graduate Students : Jennie Thomas, Michael Anderson; Undergraduate Assistants: Bryan Forward)
The statistical power of DNA a profile is directly dependent on an accurate assessment of the frequencies of the DNA markers being utilized. If a particular allele is relatively rare, a match between a suspect reference sample and a crime scene sample becomes significant. If the allele is common, a match is much less significant. Thus, the frequencies of the various alleles being tested must be rigorously known in the population in which the crime occurred (e.g. United States, Central America, Japan, etc.). Among small or isolated/inbred populations, the allele frequencies at the STR loci currently being used in forensic DNA profiling are likely to differ from those of the surrounding population as a whole. Thus, solving crimes among these populations requires the development of a separate database of allele freequencies.
In January of 2001, Dr. Ballard and one of her graduate students, Mary Hansen, traveled to Northern Tanzania to collect samples from the Masaai tribe. The Masaai consist of about 100,000 members spread over Tanzania and Kenya and are known to have experienced a genetic bottleneck in the 1800s during colonization of the area by Europeans. Since that time, the numbers of Masaai have grown, However, the population is still believed to be relatively genetically isolated.
The frequencies of the nine STR Profiler loci among the Masaai of Northern Tanzania will be determined by PCR amplification and capillary electrophoresis on a 310 Genetic Analyzer. The allele frequencies will then be compared with those of non-Masaai Tanzanians localized in the same general area (Kilimanjaro porters). If significant allele frequencies are observed, the new allele frequencies will be sent to the FBI for databasing and for use in solving future crimes among the Masaai.
A study of the allele frequencies of the same nine STR loci is also being performed among the Hispanic population of five counties in Central California to determine if the U.S. Hispanic population frequencies are appropriate for use in solving crimes involving Hispanics in this geographic area.
