Catalog  |  Cart  |  Log In

• Corporate
• Products
• Resources

Development and Validation of a GBAÔ-based mtDNA Sequence Evaluation System for Forensic Testing

In recent years, emphasis has increased for the forensic typing of poorly preserved and limited samples. Routine mitochondrial DNA (mtDNA) typing of single hairs, skeletal remains and extremely decomposed materials has been greatly facilitated by the advancements of automated DNA sequencing following PCR amplification. Although this methodology provides very precise information regarding the DNA sequence of the two hypervariable regions of the mtDNA control region, the process itself is extremely time and labor consumptive considering the low number of informative loci that are actually encountered. In an attempt to streamline testing of mtDNA sequences, specific testing of the more highly variable base positions within the control region have been developed, which include an SSO-based test strip method and a microsequencing assay. Using the flexible platform of Genetic Bit Analysis (GBAÔ) we have developed a rapid screening system for several variable mtDNA control region variants which circumvents the hybridization problems of the ASO system and equipment requirements of microsequencing, while providing an extremely robust and inexpensive test with excellent throughput potential. GBAÔ relies on dideoxy chain terminator chemistry followed by an ELISA-like colorimetric detection system. The initial development of the mtDNA array discussed here consists of ten specific base positions, mirroring the markers developed by the British Forensic Science Service for microsequencing, reporting a discriminating power (P_m) of 0.063 for Caucasians and 0.032 for Afro-Caribbeans. Additional markers, currently under development, significantly increase the value of this test on North American populations. Results of GBAÔ testing were verified through DNA sequencing, and environmental insult samples were tested to establish the robustness of the testing system. The GBAÔ system independently assays for all four of the possible nucleotides at each of the variable positions. As a result of this testing methodology, heteroplasmic sites are clearly and unambiguously identified, allowing the full value of these variants to be realized. The GBAÔ -based mtDNA testing platform allows for the rapid and inexpensive screening of mtDNA markers for routine forensic casework and has the added benefit of high throughput for sample analysis in mass disaster circumstances.

This work describes the design and implementation of a new testing platform, GBAÔ (8), for determining mtDNA haplotypes for forensic investigations. GBAÔ, named by analogy to the storage of computer data in bits, is a non-radioactive single-base sequencing method performed on a solid-phase substrate. The GBAÔ system involves the following steps; 1) PCR amplification of the target DNA sequence containing the variant base using one exonuclease-resistant primer per set of primer pairs; 2) Generation of a single stranded target by exonuclease digestion; 3) Capture of the target strand by hybridization to a pre-synthesized GBAÔ primer covalently attached to the substrate; 4) Extension of the GBAÔ primer with labeled chain terminators through single base extension; and 5) Detection of the extension product by colorimetric or direct fluorescent assay.

Organic DNA extractions were performed on 513 blood samples from Caucasians (n=103), African Americans (n=109), Hispanics ("eastern" – south Florida, n=116; "western" – south Texas, n=108), Asians (n=47), and Native Americans (n=31). Process groups of 78 individual DNA samples were prepared in 96-well plates to facilitate automated sample handling during PCR setup and GBAÔ hybridization and detection.

Ten mtDNA loci were developed for GBAÔ testing (Figure 1) that were amplified in two amplicons consisting of the HV1 and HV2 regions (Figure 1). Amplification primers corresponding to mtDNA heavy strand sequences were prepared with the first four 5’ nucleotides linked with phosphorothioated bonds. Amplifications were conducted with PE Taq DNA polymerase under typical buffer conditions and 30 cycles. Following amplification, the amplified product was digested with T7 exonuclease (5' to 3' exonuclease) added directly to the PCR reaction wells by an ICN multidrop. The enzyme digests the unprotected (non-phosphorothioated) DNA strand, yielding >99% single stranded product after 1 hour incubation at room temperature. Single stranded template DNA is then hybridized to 96-well plate bound GBAÔ capture oligonucleotides specific to the locus being interrogated. GBAÔ plates are divided into separate plates by locus each having space for 78 samples and controls. GBAÔ hybridization salts and an aliquot of the amplified DNA samples from the PCR plates was dispensed into each reaction well by a Biomek fluid handler. The GBAÔ plates also contain wells carrying control templates, reagent controls and negative controls.

Hybridization of the single-stranded PCR DNA to the capture oligonucleotide was carried out by adding an equal volume of 3M NaCl/50 mM EDTA (10ul) to the PCR reactions, and incubating the template in the presence of the plate-bound GBAÔ primer for 15 minutes at room temperature. The plate was then washed three times with TNTW buffer (10mM Tris-HCl pH 7.5, 150 mM NaCl, 0.05% Tween^® 20) using a standard microtiter plate washer (Dynatech), to remove unhybridized templates.

Locus interrogation was conducted with an extension mix containing E. coli DNA polymerase, buffer and biotinylated and fluorosceinated acyclo chain terminators was added to the washed GBAÔ plates by the Biomek. Although the majority of SNPs so far developed have displayed biallelic representation, for forensic test development each possible nucleotide is interrogated, thus two copies of each GBAÔ plate are assembled allowing for two color detection of all four bases in two parallel reactions for each SNP being evaluated.

h =(1-Sx²)n/(n-1),

where Sx² is the sum of squares of haplotype frequencies and n is the sample size (9). Estimates of the probability of a random match were calculated simply as p=Sx² following formula 2 of Stoneking et al. (1). Individuals exhibiting heteroplasmy at any of the ten loci were omitted from statistical calculations and are summarized independently.

Five hundred and thirteen individuals were tested for the ten GBAÔ loci developed in this study. Sixty-seven hapoltypes were detected among the 504 individuals that did not exhibit heteroplasmy at any locus. The frequency of the most common haplotype across all populations was approximately 20 percent. The ten-locus panel at present is capable of detecting approximately 90 percent of the genetic diversity in the regions being assessed. Statistical data, delineated by population, are found in Table 1. The results obtained with the ten locus panel described here are concordant with the data obtained by the minsequencing array of Tully et al., (6) and the overall distribution of haplotypes (Figure 2) is in agreement with population data presented by Melton et al. (3, 4) although our data represent fewer genetic loci. The inclusion of additional loci to the GBAÔ panel will increase the power of discrimination of the test to the 0.99 level desired.

Forensic validation of the GBAÔ-based kit is scheduled for completion by February of 1998. A panel consisting of 12 mtDNA loci will be tested on simulated casework samples and proficiency test materials following the guidelines of TWGDAM and DAB. Following validation studies, the GBAÔ testing platform will be organized into a kit that will be distributed to alpha and beta testing sites in the first quarter of 1998.

Table 1. Population statistics for ten mtDNA loci interrogated through GBAÔ methodology.

Go to proceedings home page